Fairness emergence from zero-intelligence agents.

Area of Science:

• Behavioral Economics
• Game Theory
• Evolutionary Dynamics

Background:

• Cooperation and fairness are crucial for social interactions.
• Previous models of fairness focused on opponent-oriented social utility.
• These models assume agents consider both their own and opponents' outcomes.

Purpose of the Study:

• To propose and investigate a payoff-oriented mechanism for fairness emergence.
• To explore if fairness can arise with minimalist assumptions, focusing solely on own payoff signals.
• To analyze the dynamics of fairness in an adaptive ultimatum game.

Main Methods:

• Utilized an adaptive ultimatum game simulation.
• Agents updated beliefs based solely on previous game payoff signals.
• Did not consider opponents' behaviors or outcomes directly in belief updates.

Main Results:

• Fairness behaviors emerged under minimalist, payoff-oriented assumptions.
• Average game payoffs decreased as the discrepancy between giving and asking rates increased.
• Belief update dynamics naturally led to a 50%-50% fair split without mutations.

Conclusions:

• Fairness preferences can emerge from simple payoff-based learning, not requiring complex social utility.
• The adaptive ultimatum game demonstrates the robustness of fairness emergence.
• Payoff-oriented mechanisms provide a parsimonious explanation for fairness in cooperative behaviors.