Theories that explain versus theories that describe: the case of constraint satisfaction
Theories, by which I mean formal theories rendered in mathematical or computational formalisms, are commonly understood as describing the data on human cognition, or as implementing a computational (in the sense of Marr) specification of cognition. Beyond these functions, we also desire theories that explain the phenomena of cognition. Three scenarios where a new theory seems to explain cognition are: (1) the development of a new mechanism that captures empirical anomalies beyond the reach of existing theories; (2) the proposal of a new theory that provides a unified account of domains previously thought to be independent; and (3) the importation of a new formalism into cognitive science that supports theories that seem to effortlessly account for large swaths of cognition. In these scenarios, a new theory explains in part by offering a new perspective on cognition, one that directly captures phenomena without much need for tedious, low-level calculation. We illustrate these claims by considering the case of constraint satisfaction theories of cognition in some depth.
thanks so much for your talk! I very much appreciate your comments about rationality and optimality. Theories can definitely help us think about these concepts much better. one thing I did not hear in your talk was any mention of simplicity. What do you think about the importance of simplicity of theories: should theories be favoured that explain m...
Thanks for the presentation! I really enjoyed your nuanced perspective on theories of cognition and styles of information processing that are differentially suitable to different goals. Are there further references you would recommend that go into more details into the constraint-based processing style that you introduced in this presentation?
Thanks Sashank, I loved your talk! I learned a lot! The talk also helped me understand my own (unreflected) intuitions about what counts as explanation better. For instance, I also feel constraints (in general, not just optimality or adaptiveness) are explanatory at the computational level, but I had not yet thought about the 'why' so explicitly. ...