Multicomponent deceptive signals reduce the speed at which predators learn that prey are profitable

Abstract

Many prey use multicomponent deceptive signals to fool predators into mistaking them for inedible objects, toxic prey, or dangerous animals. However, recent experiments have suggested that multicomponent deceptive signals are no more effective in deterring predators than single-component signals, making it difficult to understand how they have evolved. Here, we use an established experimental system in which naive domestic chicks are presented with models of snake-mimicking caterpillars to test the idea that multicomponent deceptive signals reduce the speed at which predators learn that prey are profitable. We presented chicks with a series of 4 trials in which they encountered a single type of caterpillar model. The type of model differed among our 4 experimental groups that were arranged in a 2×2 factorial design: models either possessed eyespots or did not and were in either the resting or defensive posture. Chicks’ responses to the same model prey were then retested following an extended 72-h retention period. Chicks rapidly attacked prey with no defensive traits and initially showed similar levels of wariness to prey with either 1 or 2 deceptive traits. However, chicks learned that single-trait caterpillars were profitable more quickly than 2-trait caterpillars and retained their learned responses better. This suggests that prey with multicomponent deceptive signals may have a selective advantage over prey with single-component deceptive signals when predators repeatedly encounter such prey.