Theories of Brain Function and the Nature of Vision

Our dream of distinguishing psychological functions without analyzing neuroscientific detail is unsustainable. This book presents David Marr's computational theory of vision, focusing on his separation of computational, algorithmic and implementational levels.

Over the past forty years the visual system has been the most important battleground for competing general theories of brain function. In their most interesting form, theories of brain function are directed towards answering basic philosophical questions concerning the nature of mind. A large part of this monograph involves a presentation and analysis of David Marr's computational theory of vision, focusing especially on his tripartition of computational, algorithmic and implementational levels of analysis. This historical analysis is intended to provide the motivation for a reevaluation of the central methodological presuppositions in the study of the visual system. Ultimately, this book demonstrates that the dream of being able to distinguish psychological functions without attending to neuroscientific detail has come to an end.

First detailed reading of the methodological and theoretical influence of David Marr's foundational book in cognitive science: Vision. First book length study of the development of recent theories of vision since David Marr. A detailed study of the interplay between conceptual and experimental investigations of vision from the 1981 Nobel prize winners David Hubel and Torsten Wiesel to the present.

Inhalt
Preface.- 1. Introduction.- 2. Vision as a Philosophical and Scientific Problem.- 3. The Science of Vision Before Marr.- 4. Marr's Methodology and the Computational Theory of Vision.- 5. Function/Structure and Multiple Realizability.- 6. The Anatomy of the Visual System and the Taxonomy of Visual Perception.- 7. Neural Plasticity.- 8. Ecological Theories of Vision.- 9. Conceptual Investigation: Understanding the Logic of the Visual System.- Conclusion.- Bibliography.- Name Index.- Subject Index.