Which traits do observers use to distinguish Batesian mimics from their models?

Abstract

Batesian mimicry, in which a harmless mimic resembles a more aversive model, can encompass a wide range of morphological traits, but the resemblance is never perfect. Previous studies have used abstract "prey" designs to show that differences in certain traits may not be relevant to mimicry if they are not perceived or recognized by a predator. Here, we extend these results by examining how human "predators" respond to realistic variation in traits of aposematic wasps and their hoverfly mimics. We measured the ability of humans to discriminate between images of wasps and hoverflies in which only certain traits were visible, to determine the contributions of those traits to discrimination decisions. We found that shape is a particularly useful and easily learnt trait for separating the 2 taxa. Subjects did not successfully discriminate on the basis of abdominal patterns, despite those containing useful information. Color similarity between wasps and hoverflies is relatively high in comparison with other traits, suggesting that selection has acted more strongly on color. Our findings demonstrate the importance of consideration of natural variation in the traits of prey and their salience to predators for understanding the evolution of prey defenses.