Facultative or Hidden Signals of False Infection
Until now we have treated infection mimicry as a permanent trait. However, the evolution of facultatively expressed dishonest signals of infection could be particularly beneficial. Many sickness behaviours, or behavioural symptoms such as sneezing, could be exhibited flexibly, and so could be shown to rivals, but concealed from potential mates or social partners. This flexibility would mitigate the trade-off between attraction and repulsion of different subsets of conspecifics. Situational expression of sickness behaviours mediated by the social setting has been demonstrated in some case studies (Owen-Ashley & Wingfield 2006; Lopes et al. 2012). For instance, zebra finch males who were made sick showed sickness behaviours in isolation, but not in the presence of a female or in a group (Lopes et al. 2012; Lopes et al. 2013), indicating flexibility in the expression of these behaviours based on audience composition. Sickness behaviours and behavioural symptoms are likely the most conducive to flexibility. Permanent visual signals would likely not be able to be changed in real-time, and longer-lasting signals such as scent marks on a territory (Kavaliers et al. 2005) would not be amenable to flexible deployment.
Just as flexibility in signal deployment could reduce the costs to attractiveness of mimicry, so too could taking advantage of polymorphisms in the sensory perception of receivers. Sensory systems often differ markedly among species, and even within species (reviewed in: Dangles et al. 2009). For example, in some New World primate species, males and females differ in their ability to perceive colours (Jacobs 1994). Situations such as this could allow certain signals to be perceived disparately between the sexes. Taking advantage of such differences in sensory capabilities might be especially useful in interspecific infection signalling, such as deterring predators, as mimicry in a signalling channel only available to predators would allow species to be mimics to predators without compromising cues important for intraspecific communication.
Flexible or “hidden” signals in intra-specific mimicry, however, raise some theoretical difficulties: It is difficult to reconcile why the low costs of these signals (i.e., near zero reduction in attractiveness to mates) would not result in these signals spreading to high frequency and, thus, select for receivers to ignore them. Maintenance could perhaps be achieved if there are steep cognitive costs to receiver’s ability to detect the trait, or if the signal itself were costly to produce, meaning not all individuals can generate the signal. Alternatively, what began as a signal exploiting a sensory bias against infection cues could be coopted for other purposes, such as becoming a general warning or courtship signal. The selective benefits leading to the origin of a trait (including signals) can differ from the benefits that cause a trait to be maintained or elaborated (Williams 1966; Arnqvist 2006). Thus, even if the costs of mimic signal production are low (Szamado 2003), if it is reliably deployed in specific circumstances such as agonistic interactions, then what began as an infection-mimicking signal, based on the arguments outlined above, could subsequently evolve to be part of a normal species’ repertoire of agonistic behaviours.