Natural Variations in Social Behaviors: Phenotypic Consequences and Genetic Differentiation in Paper Wasps

Abstract

Understanding how phenotypes and genotypes respond to changing environments is a central topic in evolutionary biology. Animal behavior may respond rapidly to changing environments, but little is known about the impact these behavioral responses have on the population phenotypes and genotypes. Animal recognition is an ideal system to study intraspecific variation in behaviors because signalers and receivers interact through correlated phenotypes, which allows us to test how phenotypes and genotypes interact. However, the natural intraspecific variation in recognition systems has not been deeply investigated for many types of signals. Determining the natural phenotypic and genetic variation associated with recognition systems is important because then we can better understand the evolutionary and ecological mechanisms that make recognition possible in animals. In chapter one and two, we explored the geographic variation in individual recognition in Polistes fuscatus wasps across six populations, from Pennsylvania to Michigan, USA. For individual recognition to occur in P. fuscatus, signalers must have a high diversity of within-population facial and abdominal color marks. In addition, receivers must produce a unique behavioral response toward the signaler. For signalers, we measured the within-population diversity of the facial and abdominal marks. For receivers, we experimentally tested whether receivers learn and remember the individual identities of conspecifics. We found that diversity in signaler facial and abdominal marks is different across the studied populations; some populations have a much higher diversity of marks than other populations. Receivers also show geographical variation in the ability to recognize individual identities. Two out of the six studied populations showed the ability to recognize individual identities. These results suggest that individual recognition is not a species-typical trait; rather, it is flexible and varies across populations of the same species. In chapter two, we explored the genetic structure of the six populations of P. fuscatus from chapter 1. For this analysis, we used a genome reduced‐representation approach (ddRAD-seq). First, we compared the genetic structure of the six populations. Fst values suggest that there is gene flow between the populations, and Mantel test was not positive for isolation by distance. Second, we performed a GWAS to look for candidate loci under selection given the variation of individual recognition across the populations. BayeScan approach did not find signatures of natural selection on variable loci. These results suggest that selection signatures in the genome associated with individual recognition were not detected by our approach. In chapter three, we studied how social context affects the integration of complex signals in P. fuscatus. We experimentally manipulated visual and chemical signals in workers of P. fuscatus to determine what signal nest-members use when performing nestmate recognition. We found that workers of P. fuscatus use chemical signals (cuticular hydrocarbons) but not visual signals (face and abdominal color marks) for nestmate recognition. Previous work has shown that visual signals used during social interactions mediate dominant behaviors on nests. These data suggest that P. fuscatus integrate information from multimodal signals according to certain social requirements. In conclusion, we provide empirical evidence for intraspecific variation in individual recognition. Our results suggest that gene flow between populations can generate scenarios with high phenotypic variation but low genetic differentiation between the populations. Intraspecific variation in behavior might be an underappreciated factor impacting the phenotypic and genetic diversity in social animals.