Jun Otsuka

Modern evolutionary genetics employs abstract equations to describe evolutionary change. Although the mathematical treatment has enabled rigorous predictions and hypothesis testing, it has also blurred how causal interactions at the individual level are related to a population change, resulting in a philosophical puzzle as to the nature of evolutionary process/theory. In my past works, I used causal graph theory to provide causal models underlying the standard evolutionary equations such as the breeder’s equation, Lande’s equation or population genetics equations. The well-defined causal bases of the evolutionary principles not only give a clear-cut answer to the vexed philosophical problems regarding the causal nature of evolution, selection, or fitness, but also enable one to extend the traditional genetics framework to incorporate complex mechanisms such as development (inter-phenotypic causal relationships), epigenetic inheritance, and niche construction.

Phenotype is an outcome of a complex developmental structure which by itself is a product of evolution. Recent studies in Evo-Devo have tried to understand not only how developmental structures constrain evolution, but also the way these constraints themselves evolve. My current project tackles this problem from the causal model perspective. Given a certain causal structure, can we model its short- and long-term evolutionary trajectories? Are there specific conditions that give raise to biologically important properties, such as modularity or evolvability? In addressing these questions my project tries to shed light on quantitative as well as qualitative relationships between evolution and development.

The growing amount of online resources on philosophy, such as Stanford Encyclopedia of Philosophy or PhilPapers, has generated both an opportunity to explore a philosophical topic of one’s interest in massive datasets and a need for an effective tool for that purpose. With coworkers of the InPho project (PI: Colin Allen), I worked to create an integrative platform for various types of semantic analyses on a large corpus, and used this machinery to visualize and analyze relationships among philosophers. Results from these models also suggest interesting trends or changes in topics through time. That is, it looks as if these topics “evolve.” My next goal is to understand and predict these cultural trends by applying evolutionary models.