Evolutionary stability and quasi-stationary strategy in stochastic evolutionary game dynamics

Area of Science:

• Evolutionary game theory
• Mathematical biology
• Population dynamics

Background:

• Stochastic evolutionary game dynamics are crucial for finite populations.
• A known limit shows stochastic dynamics approach deterministic replicator dynamics with large populations.
• Discrepancies exist between stochastic dynamics' limiting behavior and replicator dynamics' steady-state.

Purpose of the Study:

• To explain the paradox between stochastic and deterministic evolutionary game models.
• To introduce a new concept, the quasi-stationary strategy (QSS), for finite populations.
• To reconcile the traditional evolutionarily stable strategy (ESS) concept with finite population dynamics.

Main Methods:

• Quasi-stationary analysis of stochastic evolutionary game dynamics.
• Comparison of quasi-stationary behavior with traditional ESS and replicator dynamics.
• Time scale analysis to identify conditions for model discrepancies.

Main Results:

• A new concept, quasi-stationary strategy (QSS), is proposed for large but finite populations.
• Consistency between QSS and ESS predicts long-term replicator dynamics from stochastic behavior.
• Contradictions arise when fixation time scales exceed quasi-stationary time scales.

Conclusions:

• The study provides a framework for understanding evolutionary game dynamics in finite populations.
• QSS offers a more accurate prediction of long-term behavior than traditional ESS in certain scenarios.
• The findings clarify the relationship between deterministic and stochastic modeling approaches in evolutionary game theory.