Success-biased social learning: cultural and evolutionary dynamics

Area of Science:

• Evolutionary biology
• Behavioral ecology
• Social learning theory

Background:

• Success bias is a key social learning strategy where individuals copy successful models.
• Understanding its dynamics is crucial for explaining cultural trait transmission.
• Previous models often assumed deterministic payoffs or focused on specific contexts.

Purpose of the Study:

• To develop a general model of success-biased social learning for discrete traits with stochastic payoffs.
• To investigate the conditions under which success bias outperforms unbiased imitation.
• To analyze the evolutionary stability and invasion potential of success bias.

Main Methods:

• Developed a mathematical model for success-biased social learning with two discrete cultural variants.
• Incorporated stochastic payoffs to reflect real-world uncertainty.
• Analyzed the model's dynamics using evolutionary game theory and invasion analysis.

Main Results:

• Success bias inherently favors the adoption of rare cultural variants.
• It performs worse than unbiased imitation when the optimal variant is common and payoffs are stochastic.
• Success bias often fails to invade or replace unbiased imitation in evolutionary simulations under weak selection or stable environments.

Conclusions:

• Success bias has inherent limitations, particularly when the optimal trait is common.
• The effectiveness of success bias is sensitive to payoff stochasticity and variant frequency.
• Further research should explore the optimal strength of success bias and its prevalence across different behavioral domains.