Nonmonotonic consensus transitions in bounded-confidence dynamics on unbiased networks

Area of Science:

• Social Sciences
• Computational Social Science
• Network Science

Background:

• Opinion dynamics models explore how individual opinions evolve within a population.
• Network topology significantly influences information diffusion and collective behavior.
• The Hegselmann-Krause model is a key framework for studying bounded confidence opinion formation.

Purpose of the Study:

• To investigate the impact of network connectivity and confidence bounds on opinion dynamics.
• To construct phase diagrams classifying emergent steady states in opinion formation.
• To analyze convergence times and identify factors affecting consensus.

Main Methods:

• Simulated the Hegselmann-Krause model on sparse, unbiased networks using Wilson's algorithm.
• Systematically varied confidence level (ε) and mean degree density (μ).
• Analyzed emergent steady states, convergence times, and phase transitions.

Main Results:

• Discovered a nonmonotonic reentrant transition where higher connectivity can reduce consensus.
• Identified structural isolation as a barrier to unanimity in low-connectivity networks.
• Observed two distinct slowdowns in convergence times related to resonance and critical transitions.
• Found that finite-size effects and sparse connectivity alter dynamics and phase boundaries.

Conclusions:

• Network connectivity and confidence bounds are critical determinants of collective opinion dynamics.
• Increased connectivity does not always promote consensus; it can sometimes hinder it.
• Understanding network topology is crucial for predicting opinion convergence and fragmentation.