From the Publisher
Shows how highly distributed cooperative computation deepens our understanding of the human mind/brain and catalyzes the development of computing machinery. Using metaphors to describe the brain, distributed interactions are depicted that underlie intelligence in the human, animal, or machine (robot). Covers basic ideas about neural networks, both artificial and biological. Models are provided to show how the brain works, and the schemas which mediate our perception, knowledge, and action.
From the Inside Flap
The Metaphorical Brain 2 Neural Networks and Beyond In the early 1970s, cognitive psychology and artificial intelligence (AI) were dominated by the idée fixe of "the mind as serial computer," and researchers in both fields directed every effort at fitting the mind into the straightjacket of current computer hardware. Convinced that this approach could lead only to a dead end, Michael Arbibdrawing on his knowledge of cybernetics and the neurosciencesposited a more expansive model of the brain that combined the concepts of cooperative computation among layered neural networks with dynamic working metaphors for perception and action. Those theories, as first published in his groundbreaking The Metaphorical Brain, have since had a profound impact on brain theory and cognitive science. The Metaphorical Brain 2 was begun as a revision of the 1972 classic, but quickly grew into an independent work. This new volume offers readers a timely, indepth exploration of exciting research into the interplay between brain modeling, computer design, and Al in such areas as vision, visuomotor coordination, learning, and linguistics now being conducted by Dr. Arbib, his colleagues, and researchers around the world. Just as importantly, in The Metaphorical Brain 2, Dr. Arbib develops his theory of schemasthe dynamic processes underlying all thought and actionwhich, when combined with new insights into neural networks, forms the basis for a model that works on the principles of cooperative computation and adaptability. Schema theory is also discussed in terms of its applications in robotics and computers as well as in the study of animal brains and human psychology. Written for specialists and nonspecialists alike, this lavishly illustrated volume offers readers all relevant background information on system theory, artificial intelligence, and neurosciencewith special attention given to neural networksneeded for a firm grasp of the concepts discussed in this book. Pertinent mathematics are reserved for special sections which can be skipped over without loss of textual integrity. Students, teachers, and researchers in technological applications of neural networks, neuroscience, cognitive psychology, computer science, artificial intelligence, and the philosophy of mind will find this a stimulating and invaluable resource.