Content Follows Form: From Cognitive Architectures And Generic Tasks To Robust Intelligence

This paper addresses the issue of the basic requirements for baseline intelligent systems from a cognitive perspective. The focus here is not on optimal performance on narrow specialized tasks, but on robust performance in everyday tasks and environments that might in turn provide the basis for expert performance. Historically the focus in cognitive architectures has been on providing a set of invariant mechanisms that underlie cognitive activity. Here we focus on content rather than form, specifically on the broad generic tasks characteristic of everyday intelligence and adaptivity. We discuss three such tasks, including sequential prediction, frequency-based decision-making and system control. We describe cognitive models of those tasks and their fit to human performance, and discuss how to generalize those models. Finally, we speculate about which other tasks might share the same pervasive characteristics.

Speaker's Info

Christian Lebiere is a Research Faculty in the Psychology Department at Carnegie Mellon University. He received his B.S. in Computer Science from the University of Liege (Belgium) and his M.S. and Ph.D. from the School of Computer Science at Carnegie Mellon University. During his graduate career, he studied connectionist models and algorithms and was the co-developer of the widely used Cascade-Correlation neural network learning algorithm. Since 1991, he has worked on the development of the ACT-R hybrid cognitive architecture and was co-author with John R. Anderson of the 1998 book The Atomic Components of Thought. His main research interest is cognitive architectures and their applications to psychology, artificial intelligence, human-computer interaction, decision-making, intelligent agents and neuromorphic engineering.