The following dissertation deals with cultural evolution and information transmission within a neo-Darwinian framework. Chapter I outlines the past and present state of Evolutionary Culture Theory, including its four major sub-disciplines. Building on Dawkins' replicator/vehicle analogy, I discuss an inherent problem in operationalizing Evolutionary Culture Theory that I argue has hindered its full development into a progressive science: finding a proper unit of analysis.

Chapter II explores the problem of operationalizing the evolutionary analogy to culture in greater detail. I propose as a solution that articles and the dominant themes they transmit can serve as vehicles and replicators in a quantitative model of cultural evolution I call population semantics. I then introduce an approach called Evolutionary Citation Analysis, which involves the application of Latent Semantic Analysis to full-abstract science citation indexes and then examining citation patterns on the basis of population variables. In this way, I argue that cultural-evolutionary processes such as differentiation and transformation can be investigated by examining the impact population semantic variables have on the cumulative citation of thematic replicators.

Chapter III demonstrates quantitatively and qualitatively the proposed procedure's value as a tool for Latent Semantic Differentiation. Applying the procedure to a 4-field dataset, I show that it is able to cluster articles together on the basis of their field affiliations. I then conduct a qualitative study of the differentiated semantic themes in a dataset consisting of anthropology articles.

Chapter IV applies the approach validated in chapter III to an index of psychology articles in order to examine whether the population semantics of a field affects the reproductive success of its dominant themes. Using this approach, which I call Evolutionary Citation Analysis, I demonstrate that the density and diversity of the semantic space around articles, as well as their denominational cluster affiliations, impact the citation frequencies of their dominant themes.

Chapter V explores applications and extensions of the research, and discusses its theoretical implications. The appendix describes the methods in full.