CoDiNG - Naming Game with Continuous Latent State of Agents

2024-06-27

arXiv.org

AI Summary

This study addresses the limitation of classical opinion dynamics models in capturing the dissociation between “internal intensity” and “external expression” in real-world opinion formation. We propose a continuous–discrete hybrid opinion model. Methodologically, we introduce learnable continuous latent variables—representing individual opinion intensity—into the Naming Game framework for the first time, and integrate a cognition-inspired Bayesian updating mechanism driven by empirically grounded social network topologies. Our key contribution lies in unifying cognitive plausibility (continuous, memory-trace–based intensity evolution) with sociological interpretability (discrete public expressions), thereby relaxing the classical binary or purely discrete opinion assumptions. Experiments on multiple real-world datasets demonstrate that our model significantly outperforms baseline methods in capturing empirical phenomena—including faster opinion convergence, realistic polarization dynamics, and more accurate consensus evolution trajectories.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Constraint Satisfaction and Optimization: Mixed Discrete/Continuous Optimization

Application Category

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

Abstract

Understanding the mechanisms behind opinion formation is crucial for gaining insight into the processes that shape political beliefs, cultural attitudes, consumer choices, and social movements. This work aims to explore a nuanced model that captures the intricacies of real-world opinion dynamics by synthesizing principles from cognitive science and employing social network analysis. The proposed model is a hybrid continuous-discrete extension of the well-known Naming Game opinion model. The added latent continuous layer of opinion strength follows cognitive processes in the human brain, akin to memory imprints. The discrete layer allows for the conversion of intrinsic continuous opinion into discrete form, which often occurs when we publicly verbalize our opinions. We evaluated our model using real data as ground truth and demonstrated that the proposed mechanism outperforms the classic Naming Game model in many cases, reflecting that our model is closer to the real process of opinion formation.

Problem

• Exploring opinion formation mechanisms in social dynamics
• Extending Naming Game with continuous latent opinions
• Validating model performance against real-world data

Innovation

• Hybrid continuous-discrete Naming Game extension
• Latent continuous opinion strength layer
• Discrete layer for public opinion verbalization