All animals, from honeybees to humans, devote a portion of their daily lives to the quickly reversible state of diminished responsiveness to external stimuli known as sleep. The ubiquity of sleep across taxa reflects its importance. Sleep plays a vital role in supporting critical physiological processes by maintaining the central nervous system, supporting the immune system, and enhancing cognition. Accordingly, periods of insufficient investment in sleep can lead to severe health consequences. However, this state of reduced awareness leaves animals highly vulnerable to their predators, and presents substantial opportunity costs by precluding animals from engaging in the many other activities that are also essential to their fitness. Individuals thus navigate a series of trade-offs in making decisions about investing in sleep. In groups of social animals, individuals’ choices about when, where, and how to sleep may be further affected by the behavior of nearby conspecifics, both within, as well as outside of, their group. However, to date, most sleep research has relied on polysomnography—an exceptionally thorough, yet relatively invasive technique for measuring sleep—which has largely confined sleep research to the laboratory. This constraint obscures our understanding of how ecologically and socially relevant contexts both shape and are shaped by sleep dynamics. The established research paradigm has therefore perpetuated a substantial gap in our knowledge of one of the most universal behaviors in the animal kingdom.My dissertation makes a critical step towards filling this gap by introducing novel methods to bring sleep research into the wild, and implementing these methods to shine important light on the interdependencies between an animal’s behavioral ecology and its sleep physiology. In the course of my research, I asked and answered fundamental questions about how sleep manifests in the wild at three scales of biological organization: the individual, the group, and the population. To address these questions, I leveraged collar-mounted accelerometers to monitor the sleep patterns of wild olive baboons (Papio anubis) and validated the accelerometer-based sleep classification against high-resolution infrared videography. The chapters of this dissertation focus, respectively, on the individual, group, and population scales, each using a distinct dataset of GPS locations and accelerometry collected from the baboons that range near Mpala Research Centre in Laikipia, Kenya. These baboons’ nocturnal vulnerabilities and well-studied social dynamics, both within and between groups, make them an ideal system for studying reciprocal connections between the ecological and social environment and sleep behavior.
Chapter 1 investigates how individuals balance their physiological sleep imperative with the ecological and social pressures that render sleep costly in the wild. By analyzing the sleep duration and intensity of 26 members of a group of wild baboons, I compared the influences of homeostatic regulation, the physical location in which individuals slept, and the local social environment during the sleep period on baboon sleep behavior. I found that, in stark contrast to findings from studies in the laboratory, the recent history of sleep and activity (i.e. factors pertinent to homeostatic regulation) played a relatively minor role in shaping the sleep patterns of wild baboons. Ecological and social pressures, on the other hand, had a substantial influence on sleep. Namely, baboons slept less when sleeping in unfamiliar environments and when sleeping in proximity to more group-mates, and they did not appear to compensate for this sleep loss with higher intensity sleep. These results suggest that sacrificing sleep when in risky habitats and to interact with group-mates during the night might be adaptive for animals whose fitness depends critically on avoiding predation and maintaining strong social bonds. Beyond its immediate research findings, this chapter establishes the validity of a new method for measuring the sleep of wild animals, which I also implemented in the research described in the subsequent dissertation chapters. Furthermore, it opens an exciting new frontier of scientific exploration into the social dynamics of sleep by demonstrating a collective signature of sleep in gregarious animals.
Chapter 2 builds upon the findings of Chapter 1 with a venture into this new frontier, investigating the mechanisms and consequences of collective sleep dynamics. To shed light on how variation among group-mates and the relationships they share with each other influenced their sleep patterns, I returned to Mpala Research Centre to measure the sleep patterns of 26 members of a different group of wild baboons. I supplemented these sleep data with a complete account of the nighttime positions and movements of group members that I extracted from infrared video using computer-vision tracking of the baboons in their sleep site. Using information-theoretic and network-based diffusion analyses, I found that synchronized periods of waking during the night resulted from social disturbances of sleep that propagated primarily through the group’s affiliative network and, to a lesser extent, its dominance network. Analysis of the nocturnal movements of group members indicated that spatial proximity to an awakening group-mate during the night was not sufficient to cause an individual to become active. Thus, the spatial network was not solely responsible for guiding the propagation of wakefulness during the night, but rather the affiliative and dominance networks, independent of the spatial network, also contributed to shaping collective waking dynamics. Asymmetries in the influence that individuals have over their group-mates’ sleep patterns led socially central individuals to sleep less than their socially peripheral counterparts. These results demonstrate that the collective dynamics in a highly structured animal society create costs to sleep investment that are disproportionately borne by socially central individuals. However, this sleep loss may represent an adaptive sleep sacrifice that provides socially central individuals with more opportunities to maintain the social bonds that are essential to their higher fitness.
Chapter 3 scales the study of the social dynamics of sleep to the population level, and explores, for the first time, bi-directional feedbacks between the social environment and sleep dynamics. I analyzed the movements and sleep patterns of select members of four neighboring baboon groups to understand how encounters between distinct social groups influence sleep, as well as how sleep dynamics, in turn, influence the relationships between neighboring groups. I discovered that sharing a sleep refuge with a neighboring group impaired sleep quality. Specifically, baboons experienced shorter duration, lower efficiency, and more fragmented sleep when sharing a refuge with another group compared to when sleeping as lone groups. However, sleeping together catalyzed tolerant interactions between groups that continued well beyond the sleep period. Groups were more likely to interact and these interactions were more likely to involve extended, cohesive movements after spending the night at the same site than after sleeping at separate sites. A movement path randomization analysis confirmed that these tolerant interactions deviated from the interactions expected by chance, and revealed that they scaled up to define broader relationships of tolerance between study groups across the study period. These results suggest that sacrificing sleep in a shared sleep refuge, potentially in favor of engaging in social interactions with members of the other group sharing the refuge, may play an important role in establishing relationships between groups and, thus, shaping the social organization of a population.
Taken together, these chapters demonstrate that selectively sacrificing the vital benefits of sleep—when in risky environments, and when there may be opportunities to develop and maintain social bonds, not only with group-mates but also with individuals from other social groups—may represent key adaptations in the wild. These sleep sacrifices may have important implications not only for the fitness of an individual but also for the collective dynamics of a social group and even the social organization of an entire population. By revealing the importance of social and ecological pressures—pressures that are absent in laboratory settings—in shaping sleep behavior, my dissertation research highlights the importance of studying sleep in the wild, where the adaptive value of sleep directly reflects the complex trade-offs that have guided its evolution. This dissertation also paves the way for a new body of research that explores the fitness consequences of the strategies with which individuals navigate the sleep trade-offs demonstrated by this research. Such exploration has the potential to shed critical light on one of biology’s greatest mysteries: the evolution of sleep.