Am Nat. 2020 Nov;196(5):E127-E144. doi: 10.1086/710568. Epub 2020 Sep 11.
AbstractThe mimicry of one species by another provides one of the most celebrated examples of evolution by natural selection. Edible Batesian mimics deceive predators into believing they may be defended, whereas defended Müllerian mimics have evolved a shared warning signal, more rapidly educating predators to avoid them. However, it may benefit hungry predators to attack defended prey, while the benefits of learning about unfamiliar prey depends on the future value of this information. Previous energetic state-dependent models of predator foraging behavior have assumed complete knowledge, while informational state-dependent models have assumed fixed levels of hunger. Here, we identify the optimal decision rules of predators accounting for both energetic and informational states. We show that the nature of mimicry is qualitatively and quantitatively affected by both sources of state dependence. Associative learning weakens the extent of parasitic mimicry by edible prey because naive predators often attack defended models. More importantly, mimicry among equally highly defended prey may be parasitic or mutualistic depending on the ecological context (e.g., the source of mimics and the abundance of alternative prey). Finally, mimicry by prey with intermediate defenses corresponds to Batesian or Müllerian mimicry depending on whether the mimic is profitable to attack by hungry predators, but it is not a special case of mimicry.
摘要
物种之间的拟态是自然选择进化的最著名例子之一。可食性贝氏拟态欺骗捕食者，使它们相信自己可能受到保护，而防御性缪勒拟态则进化出了共同的警告信号，使捕食者更快地学会避开它们。然而，饥饿的捕食者可能会攻击受保护的猎物，而关于不熟悉猎物的信息的好处取决于这些信息未来的价值。以前基于能量状态的捕食者觅食行为的模型假设了完全的知识，而基于信息状态的模型则假设了固定的饥饿水平。在这里，我们确定了考虑到能量和信息状态的捕食者的最优决策规则。我们表明，拟态的性质受到这两种状态依赖性的定性和定量的影响。联想学习削弱了可食性猎物寄生拟态的程度，因为天真的捕食者经常攻击受保护的模型。更重要的是，在同样受到高度保护的猎物之间，拟态可能是寄生的或互利的，这取决于生态环境（例如，模仿者的来源和替代猎物的丰度）。最后，具有中等防御能力的猎物的拟态取决于饥饿的捕食者是否有利可图，对应于贝氏拟态或缪勒拟态，但它不是拟态的特殊情况。
