X-Message-Number: 15427 Date: Mon, 22 Jan 2001 23:43:05 -0700 From: Mike Perry <> Subject: More Computers and Brains Thomas Donaldson, #15401, says ... >Would anyone who believes that N^2 connections would represent us >completely please explain? What do you mean by "represent us completely"? We are, I think, talking about whether a computer could basically imitate a human being. I allow the computer to be probablistic (unpredictable) rather than deterministic, and of course the computer can receive inputs from the outside just like a person--why not? The computer in its "imitation" of a human brain is not necessarily going through exactly the same mental events (in a suitably isomorphic sense) as some specific brain but neither would another, initially identical brain. With these thoughts in mind, it should not be too hard to see how the time and space requirements of said computer might be no more than a polynomial of that of the brain. I've also said that the computer, to carry out its work longterm, may need more memory, which could effectively circumvent any fixed bound N on the number of neurons that are imitated. Mike Perry Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=15427