Aposematism: balancing salience and camouflage

Abstract

Aposematic signals are often characterized by high conspicuousness. Larger and brighter signals reinforce avoidance learning, distinguish defended from palatable prey and are more easily memorized by predators. Conspicuous signalling, however, has costs: encounter rates with naive, specialized or nutritionally stressed predators are likely to increase. It has been suggested that intermediate levels of aposematic conspicuousness can evolve to balance deterrence and detectability, especially for moderately defended species. The effectiveness of such signals, however, has not yet been experimentally tested under field conditions. We used dough caterpillar-like baits to test whether reduced levels of aposematic conspicuousness can have survival benefits when predated by wild birds in natural conditions. Our results suggest that, when controlling for the number and intensity of internal contrast boundaries (stripes), a reduced-conspicuousness aposematic pattern can have a survival advantage over more conspicuous signals, as well as cryptic colours. Furthermore, we find a survival benefit from the addition of internal contrast for both high and low levels of conspicuousness. This adds ecological validity to evolutionary models of aposematic saliency and the evolution of honest signalling.

Camouflage

“…Furthermore, that match need only be accurate at the spatial frequency the viewer is sampling at; in other words, as limited by their visual acuity (Caves, Brandley & Johnsen, ). This allows the same pattern to function as camouflage at a distance, but a signal when the intended viewer is close (Endler, , ; Merilaita & Tullberg, ; Bohlin, Tullberg & Merilaita, ; Barnett & Cuthill, ; Barnett, Scott‐Samuel & Cuthill, ; Barnett et al ., ,b, ,b; Barnett, Cuthill & Scott‐Samuel, , ).…”

“…If the background has an oriented texture, such as the vertical grain of bark on trees such as oak, then a background‐matching animal has to orient such that its texture matches (Sargent, ; Kang et al ., , ; Barnett et al ., ,b, ,b). This demands the ability to detect the background texture, not necessarily visually because tactile cues may suffice, and orient appropriately.…”

Diversity in warning coloration: selective paradox or the norm?

“…Alternatively, a given signal may vary depending on the position of the observer. In distance‐dependent signalling, aposematic species possess pattern elements that make them appear cryptic from afar, yet conspicuous up close (Barnett & Cuthill, ; Barnett, Scott‐Samuel & Cuthill, ). Examples include Vipera snakes (Valkonen et al, ), some butterfly larvae (Tullberg, Merilaita & Wiklund, ; Bohlin, Tullberg & Merilaita, ) and spotted skunks ( Spilogale spp.…”

Sex differences but no evidence of quantitative honesty in the warning signals of six-spot burnet moths (Zygaena filipendulaeL.)*

“…The salience of these internal contrasts can constitute an important feature of warning signals, affecting predator learning (Aronsson and Gamberale‐Stille ; Barnett et al. ). Chromatic contrast was calculated according to a widely used log version of the receptor noise‐limited Vorobyev‐Osorio color discrimination model (Vorobyev and Osorio ), which takes into account the sensitivity and abundance of each cone type (relative cone abundance taken as UV = 1, SW = 1.92 MW = 2.68, LW = 2.7 for the UVS system (Hart et al.…”

“…In addition to the metrics above, two measures of visual contrast were calculated to provide data on the perceived differences between red and black areas on the moths' forewings and hindwings. The salience of these internal contrasts can constitute an important feature of warning signals, affecting predator learning (Aronsson and Gamberale-Stille 2012; Barnett et al 2016). Chromatic contrast was calculated according to a widely used log version of the receptor noise-limited Vorobyev-Osorio color discrimination model (Vorobyev and Osorio 1998), which takes into account the sensitivity and abundance of each cone type (relative cone abundance taken as UV = 1, SW = 1.92 MW = 2.68, LW = 2.7 for the UVS system (Hart et al 2000) and V = 1, SW = 1.9, MW = 2.2, LW = 2.1 for the VS system (Hart 2002;Håstad et al 2005)), as well as the noise in the photoreceptors.…”