The Roots of International Perceptions: Simulating US Attitude Changes Towards China with LLM Agents

2025-08-12

AI Summary

This study investigates the long-term evolution of international bias—particularly U.S. public attitudes toward China—under digital-era information dynamics. Method: We propose the first macro-level public opinion evolution simulation framework grounded in large language model (LLM) agents, integrating media data crawling, neutral event extraction, fine-grained user profiling, and cognitively grounded belief-updating mechanisms. Crucially, we introduce two novel design elements: “debiased media exposure” and “adversarial agents,” enabling quantitative assessment of how shifts in information acquisition patterns drive attitude reversal. Contribution/Results: The framework successfully reproduces U.S. public sentiment trends toward China from 2005 to 2024. Empirical validation identifies media framing bias and selective exposure as primary drivers of attitudinal polarization. By offering a computationally tractable, interpretable, and agent-based modeling paradigm, this work advances the scientific understanding of transnational cognitive polarization in algorithmically mediated information environments.

Technology Category

• Knowledge Representation and Reasoning: Preferences
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarization
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

Abstract

The rise of LLMs poses new possibilities in modeling opinion evolution, a long-standing task in simulation, by leveraging advanced reasoning abilities to recreate complex, large-scale human cognitive trends. While most prior works focus on opinion evolution surrounding specific isolated events or the views within a country, ours is the first to model the large-scale attitude evolution of a population representing an entire country towards another -- US citizens' perspectives towards China. To tackle the challenges of this broad scenario, we propose a framework that integrates media data collection, user profile creation, and cognitive architecture for opinion updates to successfully reproduce the real trend of US attitudes towards China over a 20-year period from 2005 to today. We also leverage LLMs' capabilities to introduce debiased media exposure, extracting neutral events from typically subjective news contents, to uncover the roots of polarized opinion formation, as well as a devils advocate agent to help explain the rare reversal from negative to positive attitudes towards China, corresponding with changes in the way Americans obtain information about the country. The simulation results, beyond validating our framework architecture, also reveal the impact of biased framing and selection bias in shaping attitudes. Overall, our work contributes to a new paradigm for LLM-based modeling of cognitive behaviors in a large-scale, long-term, cross-border social context, providing insights into the formation of international biases and offering valuable implications for media consumers to better understand the factors shaping their perspectives, and ultimately contributing to the larger social need for bias reduction and cross-cultural tolerance.

Problem

• Model US attitude changes towards China using LLM agents
• Analyze media bias impact on international perception formation
• Simulate long-term cross-border opinion evolution with AI

Innovation

• LLM agents simulate US-China attitude evolution
• Debiased media exposure extracts neutral events
• Devils advocate agent explains opinion reversals