Phase coexistence in the fully heterogeneous Hegselmann-Krause opinion dynamics model

Area of Science:

• Social Dynamics
• Computational Social Science
• Statistical Physics

Background:

• Opinion dynamics models explore how individual opinions evolve and aggregate within social systems.
• Bounded confidence models simulate opinion formation based on agent tolerance levels.
• Heterogeneity in agents and network structures significantly impacts collective behavior.

Purpose of the Study:

• To investigate the combined effects of heterogeneous social agents and network structures on opinion dynamics.
• To explore the full phase diagram of a bounded confidence opinion model with varied agent and link properties.
• To compare model behavior in different network topologies against fully mixed populations.

Main Methods:

• Utilized a bounded confidence opinion dynamics model incorporating heterogeneous agents and social links.
• Analyzed two distinct network topologies and compared them to fully mixed populations and lattice interactions.
• Examined the phase diagram to identify emergent consensus states and dynamic transitions.

Main Results:

• Coexistence of compromising and close-minded agents leads to multiple steady states.
• Societies can reach consensus around extreme opinions, a phenomenon not observed in homogeneous systems.
• Consensus can dynamically shift from one extreme to another, driven by specific agent interactions and network positions.
• Standard order parameters may obscure critical dynamic phases, such as opinion overturning.

Conclusions:

• The interplay between heterogeneous agents and network topology can generate complex opinion dynamics, including consensus overturning.
• The presence of 'bridge' agents, willing to compromise across opinion divides, is crucial for opinion shifts.
• Findings challenge the reliability of standard order parameters in capturing nuanced social dynamics.