What do you think?
Rate this book
The Adaptive Character of Thought
This important volume examines the phenomena of cognition from an adaptive perspective. Rather than adhering to the typical practice in cognitive psychology of trying to predict behavior from a model of cognitive mechanisms, this book develops a number of models that successfully predict behavior from the structure of the environment to which cognition is adapted. The methodology -- called rational analysis -- involves specifying the information-processing goals of the system, the structure of the environment, and the computational constraints on the system, allowing predictions about behavior to be made by determining what behavior would be optimal under these assumptions. The Adaptive Character of Thought applies this methodology in great detail to four cognitive memory, categorization, causal inference, and problem solving.
290 pages, Paperback
First published June 1, 1990
2 people are currently reading
73 people want to read
About the author
John R. Anderson
60 books15 followersThere is more than one author with this name in the Goodreads' database
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
December 21, 2017
- This book is an interesting piece in the history of efforts to computationally understand and create the human mind
- The book puts rational analysis of behavior as the first prerequisite in the scientific method
- It advocates for carefully studying structural properties of the environment as the second principle of the scientific method
- The third principle is to guide mechanistic models (e.g. Marr's vision program) using prior rational analysis
- For engineering purposes, this leads us to specify a generative probabilistic (deep/reinforced) program to specify an environment (say in a game engine) and simultaneously constructing bottom up mechanisms for designing agents
Other random tit bits of interesting insights:
- Short term memory has no known rational cause. It might arise due to environmental and evolutionary constraints
- Advocates the use of category priors for integrating knowledge about objects and visual entities
- Three methods for category discovery and use: feature overlap (chair and sofa has legs), linguistic labels (words or sentences) and function (sitting is common to a chair or a tree log)
- The book puts rational analysis of behavior as the first prerequisite in the scientific method
- It advocates for carefully studying structural properties of the environment as the second principle of the scientific method
- The third principle is to guide mechanistic models (e.g. Marr's vision program) using prior rational analysis
- For engineering purposes, this leads us to specify a generative probabilistic (deep/reinforced) program to specify an environment (say in a game engine) and simultaneously constructing bottom up mechanisms for designing agents
Other random tit bits of interesting insights:
- Short term memory has no known rational cause. It might arise due to environmental and evolutionary constraints
- Advocates the use of category priors for integrating knowledge about objects and visual entities
- Three methods for category discovery and use: feature overlap (chair and sofa has legs), linguistic labels (words or sentences) and function (sitting is common to a chair or a tree log)
Displaying 1 of 1 review