What Is Mind Design?
Perspectives and things
The Turing test
Intentionality
Original intentionality
Computers
Formal systems
Automatic formal systems
Computers and intelligence
GOFAI
Interpreted formal systems
Intelligence by explicit reasoning
New-fangled Al
Connectionist networks
Embodied and embedded Al
What's missing from mind design?
Notes
Computing Machinery and Intelligence
The imitation game
Critique of the new problem
The machines concerned in the game
Digital computers
Universality of digital computers
Contrary views on the main question
Learning machines
Notes
True Believers: The Intentional Strategy and Why It Works
The intentional strategy and how it works
True believers as intentional systems
Why does the intentional strategy work?
Notes
Computer Science as Empirical Inquiry: Symbols and Search
Symbols and physical symbol systems
Laws of qualitative structure
Physical symbol systems
Development of the symbol-system hypothesis
The evidence
Conclusion
Heuristic search
Problem Solving
Search in problem solving
Intelligence without much search
Conclusion
A Framework for Representing Knowledge
Frames
Artificial intelligence and human problem solving
Default assignment
Language, understanding, and scenarios
Words, sentences, and meanings
Scenarios
Scenarios and "questions"
Questions, systems, and cases
Learning, memory, and paradigms
Requests to memory
Excuses
Clusters, classes, and a geographic analogy
Analogies and alternative descriptions
Frames and paradigms
Appendix: criticism of the logistic approach
From Micro-Worlds to Knowledge Representation: AI at an Impasse
The early seventies: micro-worlds
SHRDLU: understanding natural language
"Scene parsing" and computer vision
Learning new concepts or categories
The later seventies: knowledge representation
Frames and knowledge representation
Scripts and primitive actions
KRL: a knowledge-representation language
Conclusion
Notes
Minds, Brains, and Programs
Notes
The Architecture of Mind: A Connectionist Approach
Why brain-style computation?
The connectionist framework
Computational features of connectionist models
The state of the art
Some architecture
The scaling problem
The generalization problem
Connectionist Modeling: Neural Computation / Mental Connections
Levels of analysis: neural and mental structures
The symbolic paradigm
The subsymbolic paradigm
Semantic interpretation
The subsymbolic level
Subsymbolic computation
Subsymbolic inference and the statistical connection
Higher-level descriptions
The best-fit principle
Productions, sequential processing, and logical inference
The dynamics of activation patterns
Schemata
Conclusion
On the Nature of Theories: A Neurocomputational Perspective
The classical view of theories
Problems and alternative approaches
Elementary brainlike networks
Representation and learning in brainlike networks
Some functional properties of brainlike networks
How faithfully do these networks depict the brain?
Computational neuroscience: the naturalization of epistemology
Concluding remarks
Connectionism and Cognition
Notes
Connectionism and Cognitive Architecture: A Critical Analysis
Introduction
Levels of explanation
The nature of the dispute
Complex mental representations
Representations as "distributed" over microfeatures
Structure-sensitive operations
Learning
Reasoning
The need for symbol systems: productivity, systematicity, and inferential coherence
Productivity of thought
Systematicity of cognitive representation
The systematicity of inference
Summary
The allure of connectionism
Replies: why the usual reasons given for preferring a connectionist architecture are invalid
Parallel computation and the issue of speed
Resistance to noise and physical damage (and the argument for distributed representation)
"Soft" constraints, continuous magnitudes, and stochastic mechanisms
Explicitness of rules
On "brain-style" modeling
Concluding comments: connectionism as a theory of implementation
Conclusion
Notes
Connectionism, Eliminativism, and the Future of Folk Psychology
Introduction
Eliminativism and folk psychology
Propositional attitudes and common-sense psychology
A family of connectionist hypotheses
A connectionist model of memory
Objections and replies
Conclusion
Notes
The Presence of a Symbol
A slippery LOT
The pocket Fodor
On being more explicit
Connectionism and explicit representation
Code-fixation: its symptoms and cure
All the world's a processor
Conclusions: from code to process
Notes
Intelligence without Representation
Introduction
The evolution of intelligence
A story
Abstraction as a dangerous weapon A continuing story
Incremental intelligence
Decomposition by function
Decomposition by activity
Who has the representations?
No representation versus no central representation
The methodology in practice
Methodological maxims
An instantiation of the methodology: Allen
A second example: Herbert
What this is not
It isn't connectionism
It isn't neural networks
It isn't production rules
It isn't a blackboard
It isn't German philosophy
Key ideas
Situatedness
Embodiment
Intelligence
Emergence
Limits to growth
Dynamics and Cognition
The governing problem
Two kinds of governor
Conceptual frameworks
Morals
Three kinds of system
Three conceptions of cognition
An example of dynamical research
Is the dynamical conception viable?
Notes
Acknowledgments
Bibliography