Neural embedding of beliefs reveals the role of relative dissonance in human decision-making

2024-08-13

arXiv.org

AI Summary

Existing belief research is limited by its focus on narrow topics and reliance on subjective surveys. This work proposes a novel paradigm for belief modeling grounded in real-world online debate data. We fine-tune large language models to map thousands of social beliefs into a continuous neural embedding space—enabling, for the first time, geometric representation of large-scale human belief interactions. Embedding distance serves as a quantitative measure of cognitive dissonance, revealing statistically significant associations with decision latency and attitude shifts (p < 0.001). Methodologically, we integrate belief representation learning, high-dimensional semantic modeling, and behavioral data mining from authentic discourse. Our approach supports accurate prediction of novel beliefs (mean error < 0.12) and yields interpretable neurocognitive metrics. It establishes a computational foundation for analyzing belief structure, polarization dynamics, and decision-making processes.

Technology Category

• Knowledge Representation and Reasoning: Reasoning with Beliefs
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Natural Language Processing: Discourse, Pragmatics & Argument Mining

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Beliefs form the foundation of human cognition and decision-making, guiding our actions and social connections. A model encapsulating beliefs and their interrelationships is crucial for understanding their influence on our actions. However, research on belief interplay has often been limited to beliefs related to specific issues and relied heavily on surveys. We propose a method to study the nuanced interplay between thousands of beliefs by leveraging an online user debate data and mapping beliefs onto a neural embedding space constructed using a fine-tuned large language model (LLM). This belief space captures the interconnectedness and polarization of diverse beliefs across social issues. Our findings show that positions within this belief space predict new beliefs of individuals and estimate cognitive dissonance based on the distance between existing and new beliefs. This study demonstrates how LLMs, combined with collective online records of human beliefs, can offer insights into the fundamental principles that govern human decision-making.

Problem

• Study nuanced interplay between thousands of beliefs.
• Predict new beliefs using neural embedding space.
• Estimate cognitive dissonance based on belief distances.

Innovation

• Leverages online user debate data
• Maps beliefs using neural embedding space
• Estimates cognitive dissonance via belief distances