WHAT SHAPE ARE A GERMAN SHEPHERD'S EARS?: A TALK WITH STEPHEN M. KOSSLYN [7.15.02]

Introduction

When Stephen Kosslyn received tenure at Harvard, none of his colleagues in the Psychology Department had scholarly interests that overlapped with his, since most people were doing mathematical psychology. Prior to Harvard, during his time at Johns Hopkins, Kosslyn had become very interested in the brain and computation, which was the beginning of cognitive neuroscience. There weren't too many people thinking about such matters at that point.

Over time, many of his senior colleagues in the Psychology Department at Harvard retired or left, so he found myself in the position of being chair of several search committees, where he could nudge the program in a direction that, he believes, turned out to be a very good idea. Kosslyn chaired the committee that hired Dan Schachter, Patrick Cavanaugh, Ken Nakayama, and Alfonso Caramazza. "I tried to get Pinker, but failed on that one… for now," he says. "Most recently I chaired the committee that brought in Susan Carey and Liz Spelke. The department's gotten strong now. It's got a cohesive, underlying theme, which means that there is the potential for interaction."

The Department is currently oriented towards cognitive neuroscience. "Right now," according to Kosslyn, "it’s not very computational, which is a weakness. Computation is the language of information processing, not English, French, or any other natural language because there’s no reason to expect the kinds of concepts and distinctions captured in natural language to be appropriate for characterizing what's going on in the brain. It's different than the objects we encounter in our daily lives. Although we don't have the right version of a computational language yet—one that’s tailored for this particular machine rather than a Von Neumann machine—computation is clearly going to be the language."

— JB

(Next....)