Quantifying camouflage and conspicuousness using visual salience

Abstract

Abstract Being able to quantify the conspicuousness of animal and plant colouration is key to understanding its evolutionary and adaptive significance. Camouflaged animals, for example, are under strong selection pressure to minimise their conspicuousness to potential predators. However, successful camouflage is not an intrinsic characteristic of an animal, but rather an interaction between that animal’s phenotype and the visual environment that it is viewed against. Moreover, the efficacy of any given camouflage strategy is determined not by the signaller’s phenotype per se, but by the perceptual and cognitive capabilities of potential predators. Any attempts to quantify camouflage must therefore take both predator perception and the visual background into account. Here I describe the use of species‐relevant saliency maps, which combine the different visual features that contribute to selective attention (in this case the luminance, colour and orientation contrasts of features in the visual environment) into a single holistic measure of target conspicuousness. These can be tuned to the specific perceptual capabilities of the receiver, and used to derive a quantitative measure of target conspicuousness. Furthermore, I provide experimental evidence that these computed measures of conspicuousness significantly predict the performance of both captive and wild birds when searching for camouflaged artificial prey. By allowing the quantification of prey conspicuousness, saliency maps provide a useful tool for understanding the evolution of animal signals. However, this is not limited to inconspicuous visual signals, and the same approach could be readily used for quantifying conspicuous visual signals in a wide variety of contexts, including, for example, signals involved in mate choice and warning colouration.