X-Message-Number: 26014 Date: Thu, 14 Apr 2005 10:10:12 -0400 From: Thomas Donaldson <> Subject: To Yvan, re his brain-computer proposal Hello Mr Bozzonetti again: In my discussion of your proposed way to build a mind-computer, I pointed out several points which you seem to have forgotten. The major point, however, is quite simple: when a computer acts like a mind, TIME SHARING IS WORTHLESS. Every one of your processors is connected in a net, but each one acts differently AT THE SAME TIME. And if you propose to build a mind-computer with time sharing, it simply won't work in the real world --- which is where I believe you want it to work. I don't like repeating myself, but perhaps you simply didn't save my message. Turing did good work for his time, but he wasn't seriously trying to make a computer that acted like a brain. And in brains, precisely because time matters, our neurons work simultaneously. Not time sharing, but simultaneously. Given the number of parts even you think will likely be needed to make a mind-computer, the speed of those individual parts remains too slow. There is a fundamental problem here which doesn't come from the speed of neurons or computer processors, it comes from the simple fact that the universe itself is working everywhere simultaneously. Even very fast time-sharing processors will get caught out by that, though no so easily as we ourselves would be caught out (if we worked by time-sharing). The second problem is that of growth. Your proposed brain cannot grow new neurons (at least you haven't discussed how that could happen). If your TRACs can move their connections, that's fine, but our neurons, even in adults, grow new synapses. That is, we don't just keep a constant number of synapses N. This becomes particularly crucial because it MAY lie behind our long term memories: the dentate gyrus in our hippocampus (which deals with one kind of memories) grows new neurons, and a layer of stem cells surrounding our brain ventricles creates new neurons, which migrate to various places in our brain. (In many animals they mainly go to the olfactory system, but they don't do this in us or apes, either; and even when most go to the olfactory system in animals, some do not). As for your reading, I note a lack of books specifically on how our brains and the cells which make them up work. Have you read the Byrne and Roberts book FROM MOLECULES TO NETWORKS? Biochemistry is certainly useful in understanding brains, but you'll need much more than that. Your planned mind-computer remains light years from any computer able to work like a human mind, or better than one. Best wishes and long long life to all, Thomas Donaldson Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=26014