The Computational Evolution of Cognitive Architectures

摘要

Abstract What is human mind, and how does it work? These questions have occupied humanity since antiquity, but only a hundred years ago became a topic of rigorous scientific investigation. Cognitive architectures are a part of the long-standing effort to understand how human mind and brain operate by designing and implementing computational models in software and hardware. The field of cognitive architectures emerged at the intersection artificial intelligence and cognitive science and in less than 50 years of its existence led to creation of hundreds of projects, many of which are still being developed. In this book, we trace the evolution of cognitive architectures, their abilities, and future prospects. To do so, we analyzed over 3000 publications on 80+ cognitive architectures and hundreds more surveys, research papers, and opinion pieces spanning disciplines from philosophy and cognitive science to computer science and robotics. We aggregate our findings into broad themes, such as common components of the architectures, their organization, interaction, and relation to human cognitive abilities. Besides theory, we discuss what cognitive architectures can actually do and how to evaluate their performance. Finally, in view of recent developments in AI, we consider the future of cognitive architectures and challenges they face. We hope that this book will be useful not only to researchers who develop cognitive architectures or are otherwise involved in the field, but also to anyone interested in learning about progress towards understanding and modeling human cognition.