Table of Contents

Situated robotics, natural selection, and cultural scaffolding - the ingredients of a historical explanation: meeting the challenge of a new paradigm; symbol processing as a scientific theory of human and animal behaviour; the basic ingredients of an alternative approach; can situated activity and interactive emergence explain the intentional behaviour of human beings? travelling through the book by different routes. Computers, models, and theories: the search for a unifying calculus in computational psychology; alternative types of scientific explanation; interactive emergence. Internal representation and natural selection - epistemological and ontological strains in the cognitive approach: Dennett's formulation of the problem; attempts to reconcile internal representation and natural selection; Millikan's definition of proper function; Task domains and skill domains. Connectionism - its promise and limitations as currently conceived: analytic and synthetic connectionism; Harnad's model of symbol grounding; what do we mean by "symbols"? Smolensky's "Proper treatment"; the curious case of the horseshoe crab; the limitations of an input-output model. Scientific explanation of behaviour - the approach through formal task definition: grounding scientific paradigms; Turing's time-and-motion study of a mathematician at work; Newell and Simon's physical symbol system; Marr's concept of internal representation; Chomsky and competence theories; conclusion. Scientific explanation of behaviour - the logic of evolution and learning: liberating behaviour from behaviourist explanations; functions, goals, and purposes; why is it there? fitness and survival value; historical explanation and economic models; situated patterns of activity: the natural kinds of a historical explanation of intelligent behaviour. Toward a working definition of activity - recent developments in AI that try to come to terms with the S-Domain: the advantages of an existence proof; the concepts and methodology of autonomous agent research; an autonomous agent that builds maps by finding its way in the world; Mataric and Millikan: history of use; conclusion. An examination of alternative conceptual framework for autonomous agent research: an alternative version of Mataric's wall-following program; Horswill's implements relation; a robot that learns from emergent activity; delineating the space of possibilities. Models of behaviour selection: various aspects of the problem; economic models of choice mechanisms in artificial life; choice mechanisms in autonomous agents; choice mechanisms involving situated activity. (Part contents).

Promotional Information

Another shot across the bows of the establishment in the new cognitive revolution, Hendriks-Jansen takes us through various stances to the problems of intelligence, its evolution, engineering, and development. A broad spectrum of such perspectives is carefully considered in terms of their explanatory value, in the course of which a new agenda is unfolded which is dynamic, situated and historic in approach. It is cognitive behaviorism with an open mind. -- Brendan McGonigle, Director, Laboratory for Cognitive Neuroscience and Intelligence Systems, and Reader in Psychology and Cognitive Science, University of Edinburgh This is a provocative and wide-ranging book. It combines a radical critique of current cognitive science with constructive suggestions about other explanatory possibilities. Behavioral science, as a result, looks even more difficult than we always knew it was -- and even more interesting. -- Margaret A. Boden, Professor of Philosophy and Psychology, University of Sussex; editor, Dimensions of Creativity For 40 years the computational model of the mind has been the mainstay of cognitive psychology and AI. Now, in fields as diverse as situated robotics, ethology and infant development, this model is being discarded; it is more rewarding to describe cognition as arising from a personal history of the situated, embodied experience. In this wide-ranging and important book, Hendriks-Jansen shows through careful, detailed analysis how this perspective can form the basis for a science of behavior. Required reading for those who want to follow through the implications of these new ideas. -- Dr. Inman Harvey, Research Fellow in Evolutionary Robotics, University of Sussex In Catching Ourselves in the Act, Horst Hendriks-Jansen has produced an impressive and scholarly book. This should be recommended reading for anyone with an interest in the role of scientific explanations within cognitive science, ethology, developmental psychology, and behavior-based robotics. It is an important study, crafting a synthesis from diverse sources to arrive at an exciting and persuasive conclusion which is likely to provoke fruitful debate. -- Maja J. Mataric, Volen Center for Complex Systems, Computer Science Department, Brandeis University This book presents a thorough, clear, and penetrating treatment of the full range of important concepts and ways of thinking in modern cognitive science and work on adaptive behavior. In so doing it provides a large part of the philosophical backing needed by behavior-based and situated robotics. It shows us how we really can let go of the classical notions and to what we need to take hold of in their stead. The more people who read this book, the better will be the practice of our science of cognitive and adaptive behavior. -- Tim Smithers, Visiting Professor; Faculty of Informatics, University of the Basque Country Catching Ourselves in the Act is no less than an attempt to explain intelligence. Delightful how the author dismantles traditional views in psychology, artificial intelligence, ethology, and philosophy. But he goes beyond criticism by providing alternative explanations, drawing on recent work in situated robotics. A masterpiece in combining detailed analysis with grand theorizing. A must for any cognitive scientist. -- Rolf Pfeifer, AI Laboratory, Computer Science Department, University of Zurich

Reviews

"Catching Ourselves in the Act is no less than an attempt to explain intelligence. Delightful how the author dismantles traditional views in psychology, artificial intelligence, ethology, and philosophy. But he goes beyond criticism by providing alternative explanations, drawing on recent work in situated robotics. A masterpiece in combining detailed analysis with grand theorizing. A must for any cognitive scientist." Rolf Pfeifer, AI Laboratory, Computer Science Department,University of Zurich