To mock a Mocking bird : Studies in Biomimicry

TL;DR

This paper explores the co-evolution of mimicry signals in prey and predator learning strategies using game theory, highlighting how deception and information asymmetry influence ecosystem dynamics.

Contribution

It introduces a novel game-theoretic model of predator-prey interactions with evolving mimicry signals and learning behaviors, incorporating multi-armed bandit predators.

Findings

Prey signals evolve through predator learning and deception.

Predator strategies adapt via exploration and exploitation.

Complex phenomena emerge from co-evolutionary dynamics.

Abstract

This paper dwells on certain novel game-theoretic investigations in bio-mimicry, discussed from the perspectives of information asymmetry, individual utility and its optimization via strategic interactions involving co-evolving preys (e.g., insects) and predators (e.g., reptiles) who learn. Formally, we consider a panmictic ecosystem, occupied by species of prey with relatively short lifespan, which evolve mimicry signals over generations as they interact with predators with relatively longer lifespans, thus endowing predators with the ability to learn prey signals. Every prey sends a signal and provides utility to the predator. The prey can be either nutritious or toxic to the predator, but the prey may signal (possibly) deceptively without revealing its true "type." The model is used to study the situation where multi-armed bandit predators with zero prior information are introduced into the ecosystem. As a result of exploration and exploitation the predators naturally select the prey that result in the evolution of those signals. This co-evolution of strategies produces a variety of interesting phenomena which are subjects of this paper.