I began my career in artificial intelligence, but gradually migrated to philosophy. This research inspired an interest into the relationship between the simple models of reasoning we were developing and the (usually very different) ways that humans and animals actually solve problems. In my current work, I focus on the role of teleological attributions in psychology and ethology, especially on the relationship between learning and meaning. A failure to take learning seriously has impeded research in a number of areas. In the philosophy of representation, it has led to an emphasis on views that freeze meaning during some idealized period of learning or evolution. In comparative psychology, it has led to oversimplified dichotomies between cognition and association. And in computer science, it has led to an emphasis on brute-force search strategies that overlook the kinds of elegant simplifying assumptions that structure ecologically rational search and action. My work addresses these problems by offering approaches to representation, cognition, and knowledge representation that take the latest empirical theories of learning—especially those informed by neuroscience and ecology—as their starting point.