kspencer at iti.org (Kevin Spencer) writes:
Why do people insist upon comparing the brain, an extremely complex
dynamical system, to digital computers? Neurons are *not* transistors.
The cortex is *not* a circuit board. The "memory" of a brain is *not*
the same thing as computer "memory".
I don't think it's useful to view the brain as a digital computer, but
I *do* think it's useful to view it as a digital information
processing device, sending discrete pulses from place to place and
integrating them. The concepts of bandwidth and memory capacity make
sense for neural circuits just as much as they do for electronic
circuits, and I believe the constraints they impose will have to be
taken into account if we are ever to understand the nature of
computation in the brain.
If this seems too fuzzy, here is a well-defined operational question:
How complicated would a digital computer need to be in order to
control a robot capable of simulating human behavior, e.g. visual
pattern recognition, motor control, intelligent conversation, etc.?
-- Bill
