Self-organized division of cognitive labor

Area of Science:

• Cognitive Science
• Behavioral Economics
• Social Psychology

Background:

• Division of labor enhances group efficiency in manual and cognitive tasks.
• Cognitive mechanisms underlying emergent division of labor, especially with limited communication and multiple options, remain unclear.
• Standard utility maximization often fails to distinguish between different labor division strategies.

Purpose of the Study:

• To investigate the cognitive processes driving emergent division of labor in two-person groups.
• To model how individuals coordinate task specialization without explicit negotiation.
• To identify factors influencing efficient labor division in scenarios with multiple possible strategies.

Main Methods:

• Developed an iterative two-person game allowing multiple labor divisions but prohibiting explicit negotiation.
• Implemented the game as both a human experimental task and a computational model.
• Fitted the computational model to behavioral data from human participants.

Main Results:

• The majority of human dyads achieved an efficient division of labor in the game.
• The computational model successfully explained how players' choices evolved over rounds.
• Perceived similarity between player actions and task focal points was identified as a key guiding factor.

Conclusions:

• Emergent division of labor can be effectively guided by cognitive mechanisms like perceived action-task similarity, even with limited communication.
• The developed computational model provides insights into the interplay between individual cognition and group coordination.
• Findings have implications for improving cooperation in human groups, multi-agent systems, and human-robot collaboration.