Cognitive Science and Philosophy of Mind
Research in Cognitive Science traditionally focuses on the intersections of philosophy, cognitive psychology, computer science, and neuroscience. In particular, by integrating/synthesizing theories, experiments, and arguments from the various disciplines, cognitive scientists hope to achieve deeper insights into the nature of cognition. Many of the core issues and questions of cognitive science thus have deep philosophical relevance.
The Philosophy department at Carnegie Mellon has a long history of cognitive science research, and current research includes substantive contributions to cognitive science, as well as methodological advances and epistemological analyses that directly inform the cognitive sciences. On the substantive side, Clark Glymour has focused extensively on the problem of human causal learning, frequently through theoretical and experimental collaborations with psychologists in several branches of the University of California. His 2001 book, The Mind's Arrows, explored a range of applications of Bayesian networks in cognitive psychology and neuroscience. David Danks focuses on computational models of cognitive representations, with a particular focus on human categorization, causal learning, reasoning, and decision-making. His 2014 book, Unifying the Mind: Cognitive Representations as Graphical Models, articulates and defends a cognitive architecture in which multiple types of cognition consist of distinct processes operating on a shared representational store. More recently, he has worked on the influence of goals on different aspects of human cognition. His work includes both experimental and theoretical components, and so ranges across philosophy, psychology, and computer science. In general, the substantive cognitive science research in the department extends far beyond what is typically done in so-called experimental philosophy.
On the methodological front, Kevin Kelly has applied techniques from formal learning theory to issues of the computability of human behavior. Clark Glymour and Joe Ramsey have been engaged in a multi-year effort to develop novel techniques for extracting causal and communication structures in the brain from neuroimaging data. This work has resulted in some of the first learning algorithms that are reliable on realistic data. Relatedly, David Danks has collaborated with Sergey Plis (Mind Research Network) to develop methods for learning the true, rapid connection structures in the brain from the relatively slow, undersampled time series data collected through neuroimaging. Danks has also examined the nature and confirmation of rational models (particularly Bayesian ones), which has led to the development of a novel model of inter-theoretic relations. Wilfried Sieg has worked extensively on the connections between basic cognitive operations and the nature of computation. By turning back to Turingâ€™s original, ground-breaking works, Sieg has developed a model of computation that grounds it in the space of processes that are plausible for cognitive agents (generally), which has led to novel analyses of the computational power of brain-like structures.
In addition, both faculty and graduate students have worked on topics in the history of psychology, ranging from the emergence of psychoanalysis to the development of neuropsychology in the 19th century.