Exploring the importance of stochasticity to hybrid equilibria in a discrete signaling game.

The article investigates the evolutionary dynamics and stability of hybrid equilibria in discrete signaling games, where signalers sometimes send dishonest signals and receivers imperfectly infer signaler quality. Using agent-based stochastic simulations, the study confirms that hybrid equilibria—characterized by consonant signalers always signaling and dissonant signalers signaling probabilistically—are evolutionarily attainable and stable under a range of signal cost parameters. The simulations reveal that populations exhibit irregular oscillations around hybrid equilibria due to stochastic effects, with the amplitude of these cycles influenced by mutation rates and sizes; faster evolution in one partner can dampen oscillations in the other. Additionally, the research highlights challenges in the origin of signaling systems from nonsignaling states and suggests that such oscillatory dynamics may complicate empirical detection of hybrid equilibria in natural populations.