Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CoDiNG -- Naming Game with Continuous Latent Opinions of Individual Agents

Mateusz Nurek, Joanna Kołaczek, Radosław Michalski, Bolesław K. Szymański, Omar Lizardo

TL;DR

CoDiNG introduces a memory-informed, hybrid continuous-discrete framework for opinion dynamics by integrating the Naming Game with CogSNet-like forgetting on latent opinions. It captures how internal attitude strength evolves and translates into discrete public expressions, enabling more realistic opinion diffusion on social networks. Empirical evaluation on NetSense shows CoDiNG achieving higher F1-scores than the classical Naming Game in most topics, with best gains around moderate verbalization thresholds, though performance can vary by topic. The work advances hybrid, cognitively grounded models of opinion formation and suggests directions for parameter sensitivity and multidimensional opinion representations with memory effects.

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.

CoDiNG -- Naming Game with Continuous Latent Opinions of Individual Agents

TL;DR

CoDiNG introduces a memory-informed, hybrid continuous-discrete framework for opinion dynamics by integrating the Naming Game with CogSNet-like forgetting on latent opinions. It captures how internal attitude strength evolves and translates into discrete public expressions, enabling more realistic opinion diffusion on social networks. Empirical evaluation on NetSense shows CoDiNG achieving higher F1-scores than the classical Naming Game in most topics, with best gains around moderate verbalization thresholds, though performance can vary by topic. The work advances hybrid, cognitively grounded models of opinion formation and suggests directions for parameter sensitivity and multidimensional opinion representations with memory effects.

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.
Paper Structure (18 sections, 4 equations, 6 figures, 1 table)

This paper contains 18 sections, 4 equations, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Research
  3. Discrete Opinion Models
  4. Continuous Opinion Models
  5. Hybrid Models
  6. Preliminaries
  7. Naming Game
  8. CogSNet - Cognitive Social Network Model
  9. Introduction
  10. Model overview
  11. Forgetting functions
  12. CoDiNG - New Model of Opinions with Latent States
  13. Experiments
  14. Dataset
  15. Experimental Setting
  16. ...and 3 more sections

Figures (6)

  • Figure 1: An example of weight between two nodes throughout the time in the CogSNet network with exponential and power functions and parameters set to $\mu = 0.4$, $\theta = 0.1$, and $L = 10$ days.
  • Figure 2: An example of changing opinion in the CoDiNG model.
  • Figure 3: Survey answers distribution
  • Figure 4: Correlation between question answers
  • Figure 5: F1-score for the CoDiNG model for selected questions and tested gamma parameter values. A higher $\gamma$ value indicates a more difficult change of opinion – more frequent interactions are required. The red dashed line represents the score obtained for the classical Naming Game. The results are aggregated for all surveys.
  • ...and 1 more figures