Author ORCID Identifier

Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Arts and Sciences



First Advisor

David F. Westneat


Patterns of multilevel variation in behavior, both within and among individuals, raise challenging questions about underlying mechanisms and the selective pressures acting on them. One intriguing hypothesis is that physiology shows parallel multilevel variation, and so might represent a latent trait that integrates multilevel behavioral responses. For example, foraging acquires the fuel needed to maintain metabolism, and in turn, an individual’s metabolism affects expression of foraging, and other, behaviors. Metabolism and behavior thus might coevolve to become integrated traits. Despite the appeal of this hypothesis, numerous investigations into the link between metabolism and behavior have yielded largely equivocal results.

To better understand the ambiguous link between metabolism and behavior, I designed a mathematical model to explore the dynamics of multi-trait correlations in changing environments. My model emphasized that environmental variation in resources and mortality risk lead to a diversity of correlation structures, mimicking the context- dependent nature of empirical correlations between metabolism and behavior.

Since nearly all relevant empirical work to date has dealt with average measures of discrete metabolic states at opposite ends of an activity gradient and their link to behavior, I developed a reaction norm framework to investigate the relationship between instantaneous and contemporaneous measures of metabolism and behavior. Within this framework, metabolism is regressed onto activity and I constructed mixed-effects models to assess among-individual variation in metabolic intercepts (~ resting metabolic rate) and slopes across an activity gradient, as well as their potential covariance. My framework sheds light on the relationship between metabolism and behavior in highly relevant reaction norm space which has largely been overlooked.

Experiments using house sparrows and mice revealed consistent among- individual variation in both metabolic intercepts and slopes. These results persisted after attempts to explain sources of among-individual variation. Within my framework, the ability to detect significant among-individual variation in metabolic intercepts allows my results to join a growing body of similar findings. My results also indicate considerable variation in how individuals change metabolism with activity, raising questions about the underlying processes contributing to such variation and effects on individual energy budgets. Additionally, the link between metabolism and activity may be sensitive to other behavioral, physiological and life history correlates, reflecting different selection pressures in variable environments. I propose that individuality in metabolic reaction norms may lead to new understanding of the context-dependent nature of the links between metabolism and behavior and the processes that maintain among-individual variances and covariances.

Digital Object Identifier (DOI)

Funding Information

The study was funded in part by the United States National Science Foundation (IOS1257718) in 2017, and a Mini-Ribble (Pilot) Grant (2015, 2016, 2018) and Ribble Graduate Recruitment Award (2014) from the University of Kentucky Department of Biology.