Bergstrom, Carl T.Meacham, Frazer2018-07-312018-07-312018-07-312018Meacham_washington_0250E_18971.pdfhttp://hdl.handle.net/1773/42209Thesis (Ph.D.)--University of Washington, 2018In many contexts, animals must infer salient information about another individual indirectly by observing some other characteristic of that individual. In Chapter 1 of this thesis, a model of costly signaling is developed to investigate how stochastic signal costs influence the overall cost of communication. Chapter 2 presents a model of mate choice where females must infer from his appearance whether a potential mate will choose to be a good parent to the future offspring. Chapters 3 and 4 deal with mathematical models of anxiety disorders. These disorders affect a huge number of people and can be tremendously disabling. But it is clear that the capacity for anxiety is an evolutionary adaptation. This presents a puzzle: why has natural selection not protected us from such a common malfunctioning of an adaptation? Chapter 3 develops a model that shows how the basic information constraints inherent in the problem of learning about an environment can unavoidably cause a subset of the population to be overly sensitive to signs of danger. Chapter 4 addresses the perplexing observation that as the society of developed countries has continually become safer, anxiety has increased rather than decreased. A model is presented that shoes how the mismatch between a modern environment and the environment to which we adapted can cause this seemingly paradoxical increase in levels of anxiety. This result is in some ways analogous to the well-known ``hygiene hypothesis" of inflammatory and autoimmune diseases.application/pdfen-USnoneBehaviorEvolutionMathematical modelingPsychologyTheoretical biologyBiologyMathematicsEvolution & developmentBiologySex and Fear: Mathematical models of mate choice, parental care, and maladaptive anxietyThesis