I read On Intelligence: How a New Understanding of the Brain will Lead to the Creation of Truly Intelligent Machines by: Jeff Hawkins and Sandra Blakeslee a few weeks ago. Excluding any spiritual reading, it has been my most interesting read this year.
Jeff is the inventor of the Palm Pilot and Sandra is a science writer for the New York Times. The book is really Jeff’s baby with Sandra helping to keep the prose at a level that we can all follow.
Jeff lays out a framework to view the mind, specifically the neocortex: memory-prediction framework. Overview from Wikipedia:
The theory is motivated by the observed similarities between the brain structures (especially neocortical tissue) which are used for a wide range of behaviours available to mammals. The theory posits that the remarkably uniform physical arrangement of cortical tissue reflects a single principle or algorithm which underlies all cortical information processing. The basic processing principle is hypothesized to be a feedback/recall loop which involves both cortical and extra-cortical participation (the latter from the thalamus and the hippocampus in particular).
The memory-prediction framework provides a unified basis for thinking about the adaptive control of complex behavior. Although certain brain structures are identified as participants in the core ‘algorithm’ of prediction-from-memory, these details are less important than the set of principles that are proposed as basis for all high-level cognitive processing.
I remember exclaiming at one point in the book as I saw how Jeff’s framework was an excellent environment for memes to grow. That understanding has slipped away. I have not been able to piece back together that realization. In trying to put humpty dumpty back together Google clued me in that it was not a very original thought. Apparently Evan Louis Sheehan even has a book out on the very subject: The Mocking Memes: A Basis for Automated Intelligence. Evan states:
The book suggests that the human brain encodes a meme by a hierarchy of cortical neurons in exactly the same structure as proposed by Jeff Hawkins in his recent book On Intelligence. Hawkins’ memory-prediction model is based on Donald Hebb’s very simple learning algorithm – neurons that fire together wire together. It is a self-organizing cortical architecture that automatically captures information about all sorts of recurring real-world patterns, expressing them in hierarchies of nested relationships. As a human brain is exposed to recurring patterns, it builds bidirectional hierarchical structures upward, from sensory neurons, through levels of cortical nodes.
Each of the many overlapping hierarchies in a mature cortex represents a concept, or meme – a specific pattern that has been repeatedly observed in the environment, whether it is a simple spatial pattern, such as the shape of a tree, or a complex spatio-temporal pattern, such as a commonly performed cultural behavior. Each hierarchy is expressed as a complex set of nested relationships among patterns of sensory elements (such as photoreceptors), which are shared among all the overlapping hierarchies at the tips of their ‘roots’.
He continues to entice me with more and more promise of not just putting humpty dumpty back together but with upgrades. He definitely takes Jeff’s ideas to a whole new level. The Mocking Memes is officially on my need to read list.