Noveau neurons are better than no neurons at all
An artificial neuron built by researchers at U.C. San Diego may be the first step toward restoring brain function in patients suffering from stroke, Alzheimer's and other neurological dysfunction. The interdisciplinary team, led by Alan Selverston and Henry Abarbanel, successfully integrated their electronic neuron within a group of 14 biological neurons from the California spiny lobster.
The neural network controls the rhythmic way food passes from the stomach to the lobster's digestive system. By removing crucial biological neurons from the natural circuit, the UCSD researchers essentially created "lobster epilepsy" and then cured it. The Navy would like to emulate biological systems for information processing and control. One approach is to understand biological neural networks. "This demonstration shows that we know enough about the biology of a simple network so that we now can build models and hardware to take advantage of their nonlinear properties," said ONR Program Manager Joel Davis.
The electronic neuron was built from $7.50 worth of circuit parts from Radio Shack. The team said the key finding was in the mathematical modeling that preceded the actual experiment. They discovered they only needed to control three variables that affect a neuron's overall function, rather than the hundreds of variables involved in the detailed biological functioning of each cell. Their finding radically simplified the mathematical algorithm used to construct the circuit. The team plans to gradually replace each of the biological neurons in the network with electronic ones. If the entire electronic network can function properly, then technology will be a step closer to replacing diseased or damaged neurons in human tissue.
Dear heavens above, they're doing a Moravec Transfer on the lobster.
-- email@example.com Eliezer S. Yudkowsky http://pobox.com/~sentience/tmol-faq/meaningoflife.html Running on BeOS Typing in Dvorak Programming with Patterns Voting for Libertarians Heading for Singularity There Is A Better Way