Co-evolution of groups and opinions in an agent-based model

Area of Science:

• Social Psychology
• Computational Social Science
• Opinion Dynamics

Background:

• Traditional opinion dynamics models often overlook group structures and focus on pairwise interactions.
• Understanding how group identification influences opinion formation and polarization is crucial for explaining social phenomena like segregation.

Purpose of the Study:

• To develop and analyze a novel model for the co-evolution of opinions and group identities.
• To investigate the impact of in-group versus out-group interactions on opinion convergence and divergence.
• To explore the role of intergroup attitudes in driving opinion polarization.

Main Methods:

• Agent-based modeling approach to simulate opinion dynamics within groups.
• Modification of interaction parameters to distinguish between in-group and out-group affiliations.
• Analysis of simulation outcomes to identify key drivers of opinion polarization and group fragmentation.

Main Results:

• Attitudes towards out-groups significantly influence the degree of opinion polarization within a system.
• In-group interactions can mitigate extreme opinion shifts and promote opinion convergence within groups.
• Specific parameter settings related to out-group perception can lead to group fragmentation.

Conclusions:

• Group identification and intergroup attitudes are critical factors in opinion polarization dynamics.
• The model provides insights into the mechanisms underlying social segregation and sectarianism.
• Findings contribute to understanding the complex interplay between social identity and collective opinion formation.