- Title
- On Intelligence
- Author
- Hawkins, Jeff; Blakeslee, Sandra
- Publisher
- Owl Books (NY)
- Year
- 2004
- ISBN
- 0805078533
- Library
- N/A
Curriculum and Classification
- Subject
- Intelligence and Knowledge
- MainCurriculum
- Process Design
- SubCurriculum
- Learning and Learning Processes
- Semester
- Semester 2
Abstract
Review from Amazon.com
Jeff Hawkins, the high-tech success story behind PalmPilots and the Redwood Neuroscience Institute, does a lot of thinking about thinking. In On Intelligence Hawkins juxtaposes his two loves--computers and brains--to examine the real future of artificial intelligence. In doing so, he unites two fields of study that have been moving uneasily toward one another for at least two decades. Most people think that computers are getting smarter, and that maybe someday, they'll be as smart as we humans are. But Hawkins explains why the way we build computers today won't take us down that path. He shows, using nicely accessible examples, that our brains are memory-driven systems that use our five senses and our perception of time, space, and consciousness in a way that's totally unlike the relatively simple structures of even the most complex computer chip. Readers who gobbled up Ray Kurzweil's (The Age of Spiritual Machines and Steven Johnson's Mind Wide Open will find more intriguing food for thought here. Hawkins does a good job of outlining current brain research for a general audience, and his enthusiasm for brains is surprisingly contagious. --Therese Littleton --This text refers to the Hardcover edition.
Reviews
Kaospilots reviews of the book. Please describe the book, why is it good or why is it bad (strength/weakness), what did the book do for you, and why do you recommend it. Plus of cause - you contact information so other people can discuss this with you and get more info if needed.
Review by Peter Busch, busch@kaospilot.dk
On intelligence is a book that combines the resent brain science into a complete new paradigm for how the human brain works, and thereby what “intelligence” is.
It is a groundbreaking new point of view, which paves the way for a new understanding of learning, intelligence and the possibility of making intelligent machines.
Each neuron in the brain has – in fact – ten times or more “feed backward” synapses than “feed forward” synapses, but the brain science in general and Artificial Intelligence (AI) and Neural Network in particular has concentrated their understanding of intelligence based on the feed forward functionality, i.e. how to get information from the outside through the senses to the centers of the brain that are supposed to react to the stimulus.
Jeff Hawkins focuses on the feed backward mechanism and changes the understanding of what and how the brain works. His basic understanding is that the brain works as a memory-prediction system, and that the ability to predict, and thus react if the reality is different from the prediction is central.
His theory of the brain will influence the way we understand learning and pedagogical issues, although this is not what he writes about.
The book is useful in the Kaospilot setting in understanding learning, what expertise is, and how the brain works.