Abstract

From humble beginnings, minds in the natural world have come to take on endless forms. Yet, our best way of studying these minds has come up short. By its own lights, comparative cognitive science has repeatedly struggled to achieve the goals that it has set out for itself. Why is this, and what can be done about it? This dissertation is directed at answering these questions. It does so in two parts. The first addresses the issue of replication; both the looming threat that a replication crisis holds for comparative cognitive science as well as the broader issue of what a replication is. Here, a deflationary account of replication is defended. This view is argued to hold implications for the practice of replication, interpreting replicability as a demarcation criterion, evaluating the replication crisis across the sciences, and evaluating results of replication experiments. Moreover, this analysis holds direct consequences for the evaluation of evidence in comparative cognitive science. Most pointedly, it is not ‘mature’ enough to experience a genuine replication crisis. Nevertheless, replication experiments, when properly framed, can still make valuable contributions to the discipline. The second part builds on the analysis in the first and shows how comparative cognitive science is inhibited by what is termed the ‘validity cycle’. Here, the causes of the validity cycle are initially identified (weak theory, underspecified hypotheses, and underdetermined experiments). Then, four ameliorative proposals that have been made are introduced and their shortcomings are analyzed. These are (1) adopting a bottom-up approach, (2) breaking cognitive capacities into dimensions, (3) formalizing the theories, and (4) identifying signatures of cognitive capacities. Modified versions of each are introduced. Taken together, it is argued that these modified proposals represent the most promising way forward for the science particularly when run simultaneously in a complementary manner.