Ethological studies not only provide insights on how and why animals perform extraordinary behaviors, but they also provide an opportunity to understand how animals are disturbed by human activities in the Anthropocene. Ecotourism, a popular recreational activity, is a worldwide growing business that can be used as a tool for biodiversity conservation. However, a lack of fundamental knowledge and effective management plans for an ecotourism site may lead to an increase in anthropogenic disturbances and ultimately degrade a site’s biodiversity. In Ubon Ratchathani, Thailand, locals discovered a unique migration of freshwater shrimp known as “Parading Behavior”, an unusual type of upstream migration in which hundreds of thousands of shrimp climb out of water and walk on land to the headwater. This behavior has captured public attention and Shrimp Watching tourism has been promoted as a must-see event at Ubon Ratchathani. While striking, this behavior is poorly understood. Preliminary evidence suggests that tourists might disturb the shrimp population during migrating. Unfortunately, we know very little about the biology of, and anthropogenic threats to, the shrimp. Therefore, my dissertation aims to: (1) understand proximate and ultimate causes of the parading behavior, (2) investigate the effect of light from tourists on parading shrimp, and (3) evaluate perceptions toward the parading shrimp from locals, stakeholders, and tourists. I use the results from these behavioral studies together with an understanding of tourists, locals, and stakeholders' attitudes toward the parading shrimp to mitigate the anthropogenic disturbances and design an effective sustainable ecotourism management plan.

The pervasive perturbation of natural systems by human activities has rapidly changed the social context of many free-ranging animals, potentially reducing the efficiency of reproductive strategies, as well as the effective population size (Ne). Behavioral flexibility can be beneficial to species confronted with rapid contextual change, and the range of flexibility may ultimately influence whether a species can buy the time needed to respond adaptively to change. From the perspective of conservation management, an understanding of species' behavioral flexibility may improve predictions regarding the effects of rapid environmental change on populations, and facilitate the application of behavioral knowledge to conservation management. Fundamentally, an animal's decision-making processes are responsible for generating flexible behavioral responses, thus the lability of mechanisms underpinning decision-making influences the flexibility of behavioral responses. Here I evaluate the study of animal decision-making across scientific disciplines. I critically assess the use of animal decision-making in conservation and suggest ways in which decision theory could enhance conservation strategies. My empirical research is focused on the influence of social context on behavioral flexibility in the endangered giant panda (Ailuropoda melanoleuca). The panda is a compelling species in which to study behavioral flexibility in the conservation context, because they are solitary, and females are seasonally-monoestrus and ovulate spontaneously. While energetic constraints play a prominent role in reproductive strategies, little is known regarding their mating system or the plasticity of reproductive behavior. Pandas are behaviorally expressive, using multiple modes of signaling during courtship, however, a holistic understanding of multimodal signaling in the species is lacking. Further, although populations are depleted throughout most of their range, the influence of social context on behavior and communication has not been described. Here we show that female signaling effort is generally lower in the exclusive presence of other females, suggesting that females can modify their behavioural efforts during the pre-ovulatory period according to the prevailing social context. We also found that multimodal signaling during social interactions did not consistently evoke an immediate, discrete response from receivers. Together these findings suggest that giant pandas demonstrate a limited degree of flexible behavioral responses dependent upon the prevailing social context.

Animals are confronted with abiotic cues, conspecific signals, and signals eavesdropped from heterospecifics that create some degree of uncertainty as to the state of their world. They must reduce uncertainty to make optimal decisions. Animals can improve the accuracy with which they make decisions by combining stimuli from different modalities, a phenomenon I term “multimodal integration”. However, in some situations an animal does not benefit from this increase in accuracy and the animal does better to ignore certain stimuli. Despite a body of literature that documents such situation-dependent integration, to date, a quantitative approach to understand the conditions in which integration is favored is lacking. My dissertation develops a framework for thinking about the functional significance of integrating stimuli in multiple modalities. First, I conceptually bring together the ideas of uncertainty, costs of mistakes and prior expectation of the state of the world in order to explain why more information sources are not always better. I then present a quantitative model that parameterizes uncertainty, the costs of mistakes, prior expectations and the costs of attending to stimuli in predicting whether or not an animal should integrate sequential stimuli in different sensory modalities. The model applies to multimodal stimuli in that different levels of uncertainty can be specified for each stimulus. This feature importantly captures the property of sensory modalities to be independently affected by uncertainty. For example, the sound of noisy traffic will not likely affect your ability to see a friend walking towards you. Finally, using yellow-bellied marmots (Marmota flaviventris), I field test the extent to which three of the model’s parameters affect integration of olfactory-acoustic predator stimuli. I found that the benefit of alert behavior and uncertainty of the second (acoustic) stimulus does indeed affect integration. Overall, this dissertation establishes a foundation for a new line of inquiry into situation-dependent integration, which will help us understand the evolution of cognitive systems, communication networks, animal signals and the ways in which individuals interact with the abiotic world.

Individual behaviors related to fitness-enhancing events, like foraging and reproduction, are often shaped by internal and external factors. Of these factors, an individuals’ social environment has a dynamic effect on their behavior and this is particularly true for different types of social relationships. For example, affiliative interactions can increase an individuals’ sense of social security and improve their health, while agonistic interactions can lead to competition that alter an individuals’ access to resources. Social network analysis allows for the quantification of an individuals’ social relationships to examine whether specific aspects of sociality, such as number of social partners or centrality to a group, affect an individuals’ behavior. Social relationships do not have uniform effects across species and social systems and therefore, likely influence fitness-enhancing events in different ways. Here, I examine how an individuals’ social environment influences three important behaviors: their personality, how they solve problems, and how they may learn. In Chapter 2, I use bivariate Bayesian models to examine the co-variance between an individuals’ social network position in both their affiliative and agonistic networks and two consistent measures of anti-predator behavior: boldness, a measure of risk-taking, and docility, a measure of defensive aggression. I found that docile individuals were less involved in their agonistic network but that docility had no correlation with affiliative behavior. Boldness was not influenced by social environment. Overall, social environment impacts different anti-predator behaviors in different ways and in marmots, aggressive relationships are more important to anti-predator behavior than affiliative ones. In Chapter 3, I used structural equation models to examine how social network position indirectly affects a suite of behaviors used to solve a novel foraging problem. I found that marmots who used a diversity of behaviors were more successful problem-solvers and those who received aggression were less so. Social position indirectly impacted problem-solving success by altering the behaviors used when attempting to solve the box. Individuals who received high aggression were less persistent, less behaviorally diverse and less behaviorally selective. Thus, an individuals’ aggressive social environment alters how they approach a novel problem. In Chapter 4, I used generalized linear mixed effects models and network-based diffusion analysis to examine whether individuals learned and whether they socially transmitted foraging information on location or foraging skills. Not all yellow-bellied marmots learned to solve a novel foraging task or became more efficient at solving it with experience. While some colonies did socially transmit information on food location, they rarely socially transmitted motor skills associated with solving the foraging task. Overall, marmots are not extractive foragers and do not need to have the skills to solve complex foraging problems, but like other rodents, attract one another to locations of food. Together, these studies illustrate the value of studying social relationships using modern social network statistics.

Natal dispersal is the permanent movement of individuals from their natal home range to a new location for reproduction. Dispersal helps maintain genetic variation and is viewed as an adaptive behavior; however, the decision rules influencing dispersal may no longer be optimal in rapidly changing environments. Climate change creates mismatches between species’ life history traits, decision rules, and the environment, and this gap may be enhanced in species whose habitats are especially sensitive to rapid change. At least two processes can permit flexible responses in a changing environment: phenotypic plasticity and/or possessing sufficient additive genetic variation to permit evolution. While numerous studies investigate species’ plastic responses to altered environmental conditions, the potential to evolve when faced with long-term changes is often overlooked in ecological studies of dispersal. Here, we use the quantitative genetic mixed model, termed the ‘animal model’, to conduct a variance decomposition of female yellow-bellied marmot (Marmota flaviventer) natal dispersal. We found significant heritable variation in the propensity to disperse when using a 60-year data set, but our estimate was not substantially different from zero when we used a smaller 18-year data set that permitted us to account for known environmental effects that influence dispersal. Nevertheless, these findings illustrate the importance of phenotypic plasticity in dispersal decisions in this system and, overall, suggest that should yellow-bellied marmots experience a future mismatch with their environment, they have some additive genetic variation that may allow them to evolve a new optimal response.

The extent the evolution of sociality was shaped by multilevel selection – a theoretical framework for natural selection occurring at levels of biological organization other than the gene – is a classic debate in biology. Though common examples are focused on social behavior, we do not know if multilevel selection significantly acts on social behavioral phenotypes in the wild. For multilevel selection to contribute to evolution, social phenotypes must be variable, replicable, and have unique fitness consequences from two or more discrete levels (e.g., the individual and the group). While the individual fitness consequences of the individual social phenotype (e.g., how connected or social an individual is) have been demonstrated, the individual fitness consequences of the group’s social phenotype (e.g., the structure and pattern of all social interactions in the group) for each individual who lives in the group is largely unknown. Here I quantify individual fitness consequences of the group social phenotype, providing evidence that the individual and group social phenotypes are discrete and quantifiable levels of biological organization. Chapters 2 and 3 show how residing in more connected social groups is associated with decreased individual reproductive success but increased individual winter survival in a wild, free-living population yellow-bellied marmots (Marmota flaviventer), a harem polygynous, facultatively social, hibernating rodent with variable and genetically heritable social behaviors. Not only does the type of group an individual resides in have fitness consequences for those individuals, but the fitness consequences of the group social phenotype are different than those of the individual social phenotype (e.g., less social individuals in more connected groups experience higher winter survival). Thus, chapter 3 quantifies the independent contributions of the individual and group social phenotypes to individual fitness by exploring the two phenotypes together in a unified, multilevel selection framework. We show that multilevel selection is indeed occurring, with selection for the group social phenotype stronger than selection for the individual social phenotype. Thus, given the social phenotype is variable, heritable, and under selection in this system, this work has shown that evolution (directly or indirectly genetic) for both social position and social structure as discrete phenotypes is possible.

The United States Marine Mammal Protection Act (MMPA 1972) resulted in federal protection and enhanced the recovery of California sea lion (CSL--Zalophus californianus) populations in the United States. As an unintended consequence of their protection, sea lion populations have expanded their range, negatively interacting with every commercial fishery on the west coast of the United States, predating endangered Pacific salmonids at upriver dams, and damaging docks/fishing vessels. Conservation and management problems with California sea lions can benefit from mechanistic insights into how individuals respond to stimuli and learn about biologically important events. This dissertation provides an in depth examination into how animal learning theory can be applied to conservation issues. I apply learning mechanisms to enhance management of two conservation issues involving sea lions; depredation (removal of fish from fishing lines) and predation of endangered salmonids at the fish ladders below the Bonneville Dam. This dissertation is structured into two reviews, an empirical test of learning on deterrents on California sea lions (Zalophus californianus) interacting with fisheries, an analysis of social transmission and the impact of culling of California sea lions foraging at the Bonneville Dam. Chapter two unpacks the behavioral mechanisms involved in non-lethal deterrents and reviews their application for depredating marine mammals. This review makes the argument that Pavlovian fear conditioning can applied to deterrents to enhance success in cases of human wildlife conflict and reduce management concerns over non-target animals. Chapter three presents a field experiment on Pavlovian fear conditioning for deterrents on wild California sea lions which examines whether conditioned individuals exhibit greater avoidance compared to control animals. Chapter four reviews how learning can be applied to solve conservation problems. Social learning, for instance, functions as a multiplier, rapidly spreading undesirable wildlife behaviors through populations, suggesting that target lethal or non-lethal management strategies may be necessary. Chapter five uses social network based diffusion analysis to show that social transmission is driving foraging of endangered salmonids by California sea lions. Social transmission has implications for management measures, and we used epidemiological models of behavioral transmission to assess current and potential lethal removal strategies.

Animal personality, or consistent individual behavior, is wide spread across taxa, and is now being linked to ecology and evolutionary dynamics. Despite interest in the ecological and evolutionary consequences of personality, few studies have used a Tinbergian approach to understanding the causes and maintenance of personality. Furthermore, there is a large amount of variation within personality traits, and as evolutionary biologists, we are keenly interested in how variation is caused, develops, and is maintained within a population. My dissertation use Tinbergian principles to try and explain personality using yellow-bellied marmots (Marmota flaviventris) as a study system. First, I explore the methodology that describes personality traits by testing whether unacquainted raters could reliably assess subjects using subjective ratings. I found that raters could reliably measure subjects, and some of these measures were valid when compared to behavioral codings. I then focus on the development of personality and found that docility and boldness do not follow the same ontogenetic path. These traits become repeatable at different life stages, and this may reflect differences in stage-specific life history strategies. I also found that boldness and docility do not form a behavioral syndrome, and that this is most likely due to the differences in development. My next chapter focuses on the causes and maintenance of personality. I test three major theoretical hypotheses — growth-mortality tradeoffs, residual reproductive value, and state-dependent safety — and found no evidence for any. I did, however, find that different environmental variables differentially influence the same personality traits across contexts suggesting that selection can influence the same personality trait through different variables depending on the context. Finally, I explore the quantitative genetics of personality. To fully understand the evolution of personality, we need to know the heritability and correlations underlying these traits. I found low heritability in most personality traits with some correlations. This dissertation shows, that in marmots, personality is heavily influenced by environment and that personality is linked to life history strategy.