Three-state opinion dynamics in modular networks

Area of Science:

• Sociophysics
• Computational Social Science
• Statistical Mechanics

Background:

• Understanding opinion evolution in structured populations is crucial for social dynamics.
• Interactions within and between communities significantly shape collective opinion formation.
• The role of negative interactions and agent inflexibility in opinion dynamics remains an active research area.

Purpose of the Study:

• To investigate opinion evolution in a community-based population with both negative and positive intergroup interactions.
• To analyze the impact of inflexible agents and indecision noise on opinion ordering.
• To model and simulate opinion dynamics using coupled mean-field approximations and Monte Carlo simulations.

Main Methods:

• Development of a coupled mean-field approximation preserving community structure.
• Conducting Monte Carlo simulations to validate theoretical approximations.
• Introducing negative intergroup interactions with probability p.
• Incorporating inflexible agents and indecision noise (probability q) for opinion switching to a neutral state.

Main Results:

• Observed continuous and discontinuous order-disorder transitions under negative interactions.
• Identified nonmonotonic ordering for intermediate community strength.
• Demonstrated that modular structure leads to nonmonotonic global ordering as indecision noise (q) increases.
• Found a dual role for neutrality: aiding minority opinion spread at moderate levels but hindering it when agents are too susceptible.

Conclusions:

• Community structure and interaction types profoundly influence opinion evolution.
• Negative interactions can lead to complex phase transitions in opinion dynamics.
• Indecision noise can facilitate minority opinion adoption, but its effectiveness is contingent on the susceptibility of agents to neutrality.