Impact of centrality on cooperative processes.

Area of Science:

• Complex Systems
• Network Science
• Computational Social Science

Background:

• Modern problem-solving often involves geographically distributed teams.
• Effective communication patterns are crucial for group performance in remote collaborations.

Purpose of the Study:

• To investigate how communication network topology affects group performance in solving complex problems.
• To explore the role of betweenness centrality in agent-based models of group problem-solving.

Main Methods:

• Utilized an agent-based model to simulate group problem-solving on NK-fitness landscapes.
• Analyzed the impact of communication network structures, specifically node betweenness centrality, on task completion time.
• Agents broadcasted fitness information and adopted strategies from neighbors.

Main Results:

• For simple tasks (smooth landscapes), network topology had no significant effect on group performance.
• In easy tasks, highly central nodes were more likely to find the global maximum first.
• For difficult tasks (rugged landscapes), higher variance in node betweenness centrality correlated positively with improved group performance.

Conclusions:

• Network topology is critical for group performance on complex, rugged tasks, but not on simple, smooth tasks.
• The variance in node centrality within a communication network can enhance collective intelligence in distributed teams facing challenging problems.
• Agent cooperation and landscape specifics significantly influence individual performance on difficult tasks.