Predator decision-making shapes the dynamics and stability of mimicry systems

Mimicry is an anti-predator strategy in which prey species (the mimic) resembles an unprofitable species (the model) to deceive predators. Despite theoretical expectations for perfect mimicry, imperfect mimicry, where the mimic only partially resembles its model, is widespread in nature. To understand how imperfect mimicry can persist ecologically, we studied the effect of different predator recognition processes on the dynamics and stability of various mimicry systems. Specifically, we extended a community dynamical model that integrates optimal foraging and signal detection theories by introducing a novel abundance-dependent recognition mechanism, where predators' perception of mimic--model similarity is influenced by the relative abundance of prey types. We demonstrate that intermediate mimic--model similarity promotes stable community dynamics and increases mimic abundance in single Batesian mimicry systems. Moreover, abundance-dependent recognition leads predators to reduce attack on mimics with low morphological similarity, further contributing to system stability. Extending the framework to a multi-mimicry system, we find that Batesian and Müllerian mimics have contrasting effects: intermediate Batesian similarity continues to stabilize the system, while high Müllerian similarity provides additional protection and can offset destabilization caused by highly similar Batesian mimics. Our study offers a novel explanation for the prevalence of imperfect mimicry in nature and highlights how recognition processes shape the ecological stability of mimicry systems.