Research banner

Jenny doing fieldwork at sunsetfield viewDana and field viewJenny doing fieldwork at sunset


Sexual selection and speciation

                Does sexual selection cause speciation? This long-standing but controversial question is receiving a lot of attention currently, partly because of the special role that mate choice can play in determining gene flow. Sexual selection is thought to cause reproductive isolation when male mating signals and female preferences diversify because that can lead to sexual isolation among populations. After many years of relative obscurity this question is now in the limelight, and evidence is beginning to accumulate in its support. Yet for the most part, this evidence is limited to inferring a role for sexual selection when mating traits differ between closely related species that avoid mating with each other. We remain remarkably ignorant of how sexual selection causes reproductive isolation and when it is likely to do so. We lack answers to fundamental questions such as: Is sexual selection a primary driver of speciation, or is it limited to certain taxa or circumstances? How important is it relative to natural selection or drift? If sexual selection is involved, is it arbitrary with respect to environment or is ultimately the product of ecologically-based divergent selection? Which kinds of sexual selection play a role -- sexual selection by sensory drive, good genes, or sexual conflict? What is the genetic basis of traits that confer sexual isolation?

                Work in my lab tackles these questions directly. We investigate behavioral and ecological causes of divergence in mating traits, the genetic basis of traits involved in sexual isolation, and are using a comparative approach to evaluate the generality of early results from model systems. To address these questions we use a combination of field observations and experiments, laboratory experiments, quantitative and molecular genetics, and comparative methods. We work at the intersection of several fields, incorporating conceptual underpinnings and methodology from evolutionary genetics, evolutionary ecology, behavioral ecology, and sensory biology. This highly integrative and multilevel approach has proven powerful for uncovering the processes guiding the evolution of behavior and the processes of speciation.

                My lab uses an ideal system to study these questions -- species pairs of stickleback fish (Gasterosteus spp.) found in the postglacial lakes of British Columbia. These are extremely young species -- less than 13,000 years old -- providing a window on the speciation process. Evolutionary replication (7 species pairs of sympatric limnetic and benthic sticklebacks) allows direct experiments to test the evolutionary mechanisms involved. We also capitalize on genomic tools available for the threespine stickleback, which has been developed into a premier system for studying the genetics of adaptation.

                Male competition’s effect on speciation has been almost completely overlooked, despite its direct effect on male reproductive success and despite the prominent role other forms of competition play in speciation. Our lab has conducted several studies that we hope will change this. We have evidence that different habitats alter male competition in a manner that generates divergent selection. We have also explored the fitness landscapes generated by male competition, and find rugged landscapes with multiple fitness peaks, again creating strong disruptive selection on multiple phenotypic traits.

Back to top


Genetics and genomics of speciation

Spawning

                This work is designed to understand the genetics of species differences in traits that confer sexual isolation in sticklebacks. The very recent origin of the stickleback species allows us to investigate the genetic changes that occur during the process of speciation, especially exciting as a counterpoint to work on the genetics of speciation in model taxa (e.g., Drosophila) where species are millions of years old and there is extensive intrinsic genetic isolation. Moreover, much work on the genetics of speciation ignores the evolutionary causes of genetic change. Our work starts by identifying those causes and their effect on phenotypes, and then explores the way this shapes the genome. We are characterizing the genetic architecture of traits known to isolate the species, such as male nuptial color, body size, and shape, as well as behavioral traits. We use a combination of quantitative genetics, QTL mapping, and admixture mapping to achieve these goals. This work is relevant to testing several models of speciation, and will lay the foundation for further work investigating sexual selection and speciation. Our eventual aim is twofold: to find the genomic regions responsible for mating trait differences, and explore how selection has acted on those regions to structure the genome.

                We have used admixture mapping to uncover the genetic architecture of divergent adaptation and reproductive isolation. This innovative approach takes advantage of hybridization that occurs between species in the wild to find genomic regions that underpin species differences. Historically, the benthic–limnetic species pair from Enos Lake showed strong divergence in many phenotypes, including male nuptial coloration and body shape, two traits known to be important in speciation. However, recent ecological disturbance caused extensive hybridization resulting in a single, admixed population. Using admixture mapping we identified genomic regions associated with divergence in male nuptial color and also body shape, suggesting that either tight linkage or pleiotropy has facilitated speciation in benthic–limnetic species pairs.

                We are also mapping fitness through sexual selection arising from both male competition and female choice. Here we use a naturalized setting to measure sexual selection on multiple behavioral and morphological phenotypes – an experiment that takes an enormous effort, but is a lot of fun! Using RAD-seq data we then do genetic mapping to find genomic regions that underpin multiple phenotypic traits, and multiple fitness components.

Back to top


Reverse speciation

               We have a rare opportunity to gain insight into the speciation process by watching it go in the reverse direction. Ecological speciation is a powerful mechanism of diversification. But species that diverge because of differences in ecology are vulnerable to ecological change. We are witnessing a natural experiment - ecological disturbance has triggered the breakdown of a species pair through hybridization. To try to find the silver lining in a sad event (the limnetic-benthic species pairs are listed as endangered in Canada), my lab is using this reverse speciation to understand the forward process of diversification. We're investigating the loss of both premating and postmating isolation – characterizing what HAS been lost and identifying the causes. This work has important implications both for basic science (speciation processes) and conservation (the effects of invasive species on diversity). To this end, we hope to facilitate recovery of this pair, and to help avoid similar problems for the other pairs.

               The silver lining may be even brighter, as hybridization can have both destructive and creative effects on diversification. A new research direction explores this tension between the homogenizing effect of hybridization and its potential to generate novel phenotypes and genotypes that could serve as the foundation for subsequent adaptation and diversification. Intriguing but controversial, the potential for hybridization to foster diversity is seen in some systems (Heliconius butterflies for one) but whether this is an anomaly or more widespread is very uncertain. The hybridizing Enos fish may give us a window on this process.

 Back to top


Sexual signal evolution

                My lab is collaborating with Tom Getty and Chris Klausmeier at the Kellogg Biological Station on a set of projects investigating the dynamics of rapid evolutionary change in sexual signals - traits like frog calls and bird plumage that are typically used by males to attract and secure mates. Our work highlights the loss of mating signals rather than their elaboration, and we use both empirical studies in stickleback fish and field crickets, and mathematical modeling to understand the evolutionary loss of these important traits. To date, most research on sexual selection has emphasized the exaggeration of male signals and very strong female preferences for those signals. Both of these should reduce diversity in sexual signaling systems resulting in one biggest, best male signal. Yet, mating signals are extremely diverse and recent studies in organisms ranging from insects to vertebrates suggest that male sexual traits are frequently lost, despite the contention that female preferences should maintain them. We hypothesize that the dynamics of both trait loss and elaboration contribute to the diversity of sexual systems we see in nature. But, why and how are male signals lost?

Back to top


Cognitive and sensory evolution


MRI images of stickleback brains showing olfactory lobe differences in L&B fish

                The lab has investigated many aspects of learning, cognition, and sensory evolution in sticklebacks. This is a great way for me to stay connected to my prior work with bats on social and vocal learning and cooperative social behavior, and to do so in the experimentally tractable and smart (for fish) sticklebacks. We’ve had many projects along these lines. Some highlights include the role of social experience in shaping mate preferences and enhancing sexual isolation. Imprinting has particular importance in this regard – without it sexual isolation is weak at best. An exciting result from the imprinting work was to identify the developmental stage (very early) and sensory cues that form the basis of imprinting – this reveals a key role for odor. Combining olfactory and visual communication in a multimodal manner may prove to be essential to speciation. The sources of variation in both visual and olfactory cues, and the causes of species differences are an active area of our current research.

                Social learning is another focus of our work. We are particularly interested in comparing social learning between stickleback species and relating this to differences in ecology and evolution in social behavior. We are also investigating the changes in stickleback brains that accompany alterations in sensory modality and cognitive ability. Lastly, we are connecting our work on mate choice, cognition, and genomics by exploring gene expression in the brain under different social conditions.

 Back to top


Evolutionary robotics

                We are collaborating with an electrical engineer (Xiaobo Tan) and computer scientist (Phil McKinley) at MSU to advance our understanding of both behavioral evolution and robotics. We plan to do this by creating realistic two-way interactions between live and robotic fish, designed within an evolutionary computational framework. We focus on predator inspection, an adaptive antipredator behavior, to study sophisticated non-cooperative and cooperative behaviors in fish, and to synthesize control and learning strategies for autonomous robotic systems operating in noisy and unstructured environments. This work is a real meeting-of-the-minds, and requires concerted, interrelated research effort in biology, robotics, and computing. Evolutionary computing will enable the design of realistic robotic fish, and the exploration of important evolutionary questions. The insight from biology will be exploited to enable adaptive behavior for autonomous robotic systems, and the use of robots will give us unprecedented flexibility to test behavioral hypotheses.

 Back to top



Jenny resting from fieldworkJocelyn and Danastickle measure

Back to top