Faculty

David Danks
Department Head

Professor of Philosophy and Psychology

Department of Philosophy

135 Baker Hall

412.268.8047

ddanks@cmu.edu

Research Interests


Most of my research involves what is sometimes called computational cognitive science: developing fully-specified computational models to describe, predict, and most importantly, explain human behavior. I have principally focused on the cognitive domains of causal learning/inference, and concept acquisition/application, but have recently begun to look at decision-making as well. I tend to do this work from within the framework of graphical models, though I am also very interested in how these high-level learning/representations/inferences could be implemented or instantiated in psychologically plausible lower-level models.

My research on causal learning has largely been inspired by two questions: (1) How could a cognitive agent (whether human or artificial) learn about causal relationships in the world from various kinds of information? and (2) How can the learned causal structure be used for inference and prediction? My research on the problem of concept learning and categorization has focused on analogous questions: How could a cognitive agent—either human or artificial—acquire meaningful categories, apply them to novel individuals, and use those categorization judgments for inference and prediction? Although these two domains have traditionally been studied in relative isolation, they are deeply relevant for one another. Thus, much of my work aims to integrate theories and experiments in the two domains.

In addition to this work directly contributing to the cognitive sciences, I have recently begun a more philosophical project to examine the explanatory power of, and evidential burdens facing, psychological theories that focus on the optimality or rationality of behavior. This issue is particularly pressing in the case of Bayesian models of cognition, which have experienced enormous popularity in recent years. In contrast to much of the rhetoric (pro and con), I have argued for a more nuanced view of these models.

Publications and C.V.

[Link]

Teaching


Undergraduate
80-150: Nature of Reason
80-257: Nietzsche
80-270: Philosophy of Mind
80-271: Philosophy & Psychology
80-316/616: Probability & Artificial Intelligence
80-323/623: Philosophy of Biology

Graduate
80-602: Proseminar/Philosophical Foundations seminar
80-514: Seminar on Philosophy of Science seminar (past seminars have included: Current Topics; Graphical Models in Cognitive Science; and Normativity in Cognitive Psychology)

 

Φ Back to Faculty Listing