Amazon Book Description
- Parallel distributed processing is transforming the field of cognitive science. Microcognition provides a clear, readable guide to this emerging paradigm from a cognitive philosopher's point of view. It explains and explores the biological basis of PDP, its psychological importance, and its philosophical relevance.
- Starting with a survey of the assumptions and methodology of classical AI Microcognition proceeds to a full treatment of the PDP alternative. It describes the main properties of PDP architectures with numerous examples and it explores the tangled question of their relationship with classical work, the theoretical significance of mixed PDP and classical models, and various criticisms of the models.
- Biological and evolutionary perspectives are also included. Of particular philosophical interest is the author's treatment of the relation between these computational models and ordinary talk of beliefs and desires, or folk-psychology.
Back Cover Blurb
- Parallel distributed processing is transforming the field of cognitive science. Microcognitionprovides a clear, readable guide to this emerging paradigm from a cognitive philosopher’s point of view. It explains and explores the biological basis of PDP, its psychological importance, and its philosophical relevance.
- “Witty and eloquent.... Definitely worth reading."
→ Amahl Smith, Times Literary Supplement
- “Andy Clark is surely one of the brightest, wisest, wittiest philosophers working in this fast-growing field today. His examples are apt, and his writing is clear and vivid. The book is as authoritative as its tone suggests.”
→ Daniel Dennett, Tufts University
- Andy Clark is a philosopher working in the School of Cognitive and Computing Sciences at the University of Sussex
Preface
- The subtitle is enough to put anyone off: Philosophy, Cognitive Science, and Parallel Distributed Processing. It might have read: the elusive, the ill-defined, and the uncharted. For all that, the project thrust itself upon me with an unusual sense of its own urgency. Parallel distributed processing (PDP) is an exciting and provocative new movement within cognitive science. It offers nothing less than a new kind of computational model of mind. The bad news is that it as yet offers nothing more than a hint of the nature and power of such models. But the hints themselves are remarkable and have the potential, I believe, to reshape both artificial intelligence (Al) and much of the philosophy of mind. In particular, they offer a new picture of the relation between sentences ascribing thoughts and the in-the-head computational structures subserving intelligent action. The final product of such reshaping will not be found in this work. At best, I offer some views on the central points of contrast between the old shape and the new, a personal view on most of the major issues, and along the way, a reasonably detailed taxonomy of features, distinctions, and subprojects. The taxonomy, though necessarily individualistic, may be of some use in future discussions of what is, in effect, a whole new topic for philosophy and Al. The conclusions are often provisional, as befits discussions of an approach that is still in its infancy. By the time this book sees print, there will be many new and relevant developments. I hope the book will provide at least a framework in which to locate them.
- The reader should be warned of my peculiar circumstances. I am first of all a philosopher, with only a secondary knowledge of Al and evolutionary biology. I am fortunate to work in the highly interdisciplinary School of Cognitive and Computing Sciences at Sussex University. It is only thanks to that harsh selective environment that I have been able to avoid many glaring misunderstandings. Those that remain are, in the time-honored clause, entirely my own responsibility. Adaptation, even to an environment of Al researchers and cognitive scientists, falls somewhat short of an optimizing process.
Contents
- Preface – xi
Acknowledgments – xiii
Introduction: What the Brain's-Eye View Tells the Mind's-Eye View – 1
- After the Goldrush – 1
- Parallel Distributed Processing and Conventional Al – 2
- The Multiplicity of Mind – 2
- The Mind's-Eye View and the Brain’s-Eye View – 3
- The Pate of the Polk – 5
- Threads to Follow – 6
- PART 1: The Mind's-Eye View
- Chapter I: Classical Cognitivism – 9
- Cognitivism, Life, and Pasta – 9
- Turing, McCarthy, Newell, and Simon – 9
- The Physical-Symbol-System Hypothesis – 11
- Bringing Home the BACON – 13
- Semantically Transparent Systems – 17
- Functionalism – 21
- Chapter 2: Situation and Substance – 25
- Stuff and Common Sense – 25
- The Dreyfus Case – 25
- It Ain't What You Know; It's the Way You Know It – 27
- Manipulating the Formal Shadows of Mind – 30
- Showing What Were Made Of – 32
- Microfunctionalism – 34
- Chapter 3: Folk Psychology, Thought, and Context – 37
- A Can of Worms – 37
- A Beginner's Guide to Folk Psychology – 38
- The Trouble with Folk – 39
- Content and World – 42
- Interlude – 46
- Some Naturalistic Reflections – 47
- Ascriptive Meaning Holism – 48
- Churchland Again – 50
- Cognitive Science and Constitutive Claims – 54
- Functionalism without Folk – 58
- Chapter 4: Biological Constraints – 61
- Natural-Born Thinkers – 61
- Most Likely to Succeed – 61
- Thrift, the 007 Principle – 63
- Gradualistic Holism and the Historical Snowball – 66
- The Methodology of MIND – 74
- PART 2: The Brain's-Eye View
- Chapter 5: Parallel Distributed Processing – 83
- PDP or not PDP? – 83
- The Space between the Notes – 84
- The Jets and the Sharks – 86
- Emergent Schemata – 92
- Distributed Memory – 96
- Biology Revisited – 104
- Chapter 6: Informational Holism 107
- In Praise of Indiscretion – 107
- Information Holism in a Model of Sentence Processing – 107
- Symbolic Flexibility – 111
- Grades of Semantic Transparency – 114
- Underpinning Symbolic Flexibility – 118
- PDP and the Nature of Intelligence – 121
- An Equivalence Class of Algorithms – 124
- Chapter 7: The Multiplicity of Mind: A Limited Defence of Classical Cognitivism – 127
- Of Clouds and Classical Cognitivism – 127
- Against Uniformity – 128
- Simulating a Von Neumann Architecture – 131
- A Lacuna in the Account of Real Symbol Processing – 136
- Full Simulation, Intuitive Processing, and the Conscious Rule Interpreter – 137
- BACON, an Illustration – 139
- Chapter 8: Structured Thought, Part 1 – 143
- Weighting for Godot? – 143
- The Systematicity Argument – 144
- Systematicity and Structured Behavior – 146
- Cognitive Architecture – 150
- Two Kinds of Cognitive Science – 152
- Grammar, Rules, and Descriptivism – 154
- Is Naive Physics in the Head? – 157
- Refusing the Syntactic Challenge – 160
- Chapter 9: Structured Thought, Part 2 – 161
- Good News and Bad News – 161
- The Past Tense Acquisition Network – 161
- The Pinker and Prince Critique – 165
- Pathology – 168
- And the Moral of the Story Is... – 169
- The Theoretical Analysis of Mired Models – 172
- Chapter 10: Reassembling the Jigsaw 177
- The Pieces 177
- Building a Thinker – 178
- Explaining a Thinker 180
- Some Caveats 183
- Epilogue
- The Parable of the High-Level Architect – 185
- Appendix: Beyond Eliminativism 187
- A Distributed Argument – 187
- Levels of Description of Connectionist Systems – 188
- Explanation Revisited – 196
- The Value of High-Level Descriptions – 199
- Self-Monitoring Connectionist Systems – 202
- A Double Error – 207
- Notes – 209
Bibliography – 213
Index – 221
Book Comment
MIT Press (1 Jan. 1989). Paperback.
Text Colour Conventions (see disclaimer)- Blue: Text by me; © Theo Todman, 2025
- Mauve: Text by correspondent(s) or other author(s); © the author(s)