Context matters: sexual signaling loss in digital organisms

Abstract

Sexual signals are important in attracting and choosing mates; however, these signals and their associated preferences are often costly and frequently lost. Despite the prevalence of signaling system loss in many taxa, the factors leading to signal loss remain poorly understood. Here, we test the hypothesis that complexity in signal loss scenarios is due to the context-dependent nature of the many factors affecting signal loss itself. Using the Avida digital life platform, we evolved 50 replicates of ∼250 lineages, each with a unique combination of parameters, including whether signaling is obligate or facultative; genetic linkage between signaling and receiving genes; population size; and strength of preference for signals. Each of these factors ostensibly plays a crucial role in signal loss, but was found to do so only under specific conditions. Under obligate signaling, genetic linkage, but not population size, influenced signal loss; under facultative signaling, genetic linkage does not have significant influence. Somewhat surprisingly, only a total loss of preference in the obligate signaling populations led to total signal loss, indicating that even a modest amount of preference is enough to maintain signaling systems. Strength of preference proved to be the strongest single force preventing signal loss, as it consistently overcame the potential effects of drift within our study. Our findings suggest that signaling loss is often dependent on not just preference for signals, population size, and genetic linkage, but also whether signals are required to initiate mating. These data provide an understanding of the factors (and their interactions) that may facilitate the maintenance of sexual signals.

When Does Sexual Signal Exploitation Lead to Signal Loss?

“…However, it is important to note that the link between genetic variation and signal or trait loss may not be so clear because genetic variation, which should be greater in large populations, is necessary to provide the fodder for selection to act on during the establishment of novel traits. In Weigel et al's (2015) modeling study, population size did not affect the likelihood of signal loss becoming established in a population, though greater variability was introduced when populations were small. This argues for a more nuanced view on the role of population size and genetic variation during signal or trait loss, and more empirical evidence from natural populations that have experienced trait loss.…”

“…We bolster these suggestions by drawing on other empirical and theoretical work on sexual signal and trait loss. A recent modeling paper (Weigel et al, 2015) was particularly useful in this regard, as the authors explicitly modeled sexual signal loss under a number of scenarios using digital organisms evolved for many generations under different strengths of female preference, flexibility of the mating system (facultative or obligate signaling), population size, and genetic linkage between preference and signal.…”

“…Female preference for a sexual signal has been implicated as a driver of signal loss in Weigel et al (2015) and some, but not all, comparative studies. Weigel et al (2015) found that it was very difficult to achieve complete sexual signal loss while females expressed any preference for the sexual signal. The only female preference condition under which sexual signal loss spread to fixation was when female preference was completely absent.…”

“…While eye loss in cavefish is the product of around 12-15 loci (Protas et al, 2007), the repeated evolution of albinism in cavefish populations is caused by independent mutations in a single gene (Protas et al, 2006). Though evolutionary change in signals should require genetic shifts in both the signaler and the receiver, Weigel et al (2015) found that genetic linkage between the signaler and receive had little bearing on whether a signal was ultimately lost. This is probably because other factors had a much bigger influence, namely whether the signal was required to initiate mating and whether females exhibited any preference (Weigel et al, 2015).…”

“…Though evolutionary change in signals should require genetic shifts in both the signaler and the receiver, Weigel et al (2015) found that genetic linkage between the signaler and receive had little bearing on whether a signal was ultimately lost. This is probably because other factors had a much bigger influence, namely whether the signal was required to initiate mating and whether females exhibited any preference (Weigel et al, 2015).…”