Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!usc!wuarchive!bcm!dimacs.rutgers.edu!aramis.rutgers.edu!athos.rutgers.edu!nanotech From: RDEES@umiami.ir.miami.edu (Matthion) Newsgroups: sci.nanotech Subject: Re: Is uploading suicide? Message-ID: Date: 15 Jan 91 02:29:16 GMT Sender: nanotech@athos.rutgers.edu Lines: 66 Approved: nanotech@aramis.rutgers.edu In article , jgsmith@bcm.tmc.edu (James G. Smith) writes: > Creating a perfect neuronal map of a person's brain on a cellular scale > (this neuron connects to these neurons, etc.) will not be sufficient to > duplicate the personality represented by that brain. Different neurons use > different molecules to transmit signals. As far as I know, one neuron may > be connected to 10 others, but only 3 of those others may be listening. Just a note: a single neuron could be connected to as many as 60,000 others. > Also, the degree of 'listening' depends on the number of receptors for > the neurotransmitter (i.e. some may be listening more closely than others). > Also, personality may be affected by hormones. (Remember PMS?) > > In order to duplicate a brain, you will have to know the composition down > to the molecular level. I find it unlikely we will be able to do that > until the very final stages of nanotechnology. The effects of hormones and "listening" can all be seen as the temporal adjustments to whatever model you are using for the brain. There is no reason that the model would have to simulate or copy the _actual_ structure. I think it would be sufficient to simply copy the temporal and "signal strength" characteristics of the brain's communications, and this could be accomplished through many mechanisms that are not directly related to neuroanatomical structure or hormonal process. It is only neccessary to capture the essential characteristics of these "nerve signal modulators." As for the whole issue of uploading as suicide, it all seems to come down to whether you believe that the human soul (ruach, pneuma, spiritis, electro- magnetic resonance, acquired personality of a human machine, whatever...) can exist if implemented on a different substrate. A hard-core materialist who can believe in AI should have no problems. One who doesn't believe in AI will probably not be convinced, and with good reason (from their perspective). Questions: (1) can a machine support consciousness? (2) if so, _must_ this consciousness arise spontaneously or could it be placed into the machine? Undoubtedly much more debate will go before the technology hits us, and gives a reason for the mental exercise. -- =========================================================================== (__) | Matthew Augustus Douglas Turner ^^ (oo) | ^^^^ /-------\/ | Department of Somthing or Other ^^^^^ / | || | College of Arcane Arts (A.A.) ^^^^^ * ||----|| | The University of Miami ^^^^^^^^ ====^^====^^==== | ^^^^^^^^^^^^^/ ^^^^ | rdees@umiami.ir.miami.edu ^^^^^^^^^^^^^^^^^^^^^^^^^^ | ...and elsewhere ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | Cow Hanging Ten at Malibu | Analytical Engines Incorporated ============================================================================ [Brain structure has a host of subtleties, and I for one wouldn't upload onto a machine built with some of the simplifying assumptions you make. An example: Until recently it was assumed that "signal strength" at a given synapse was more or less a one-dimensional function of firing rates, but within the last year or so, there have been discovered some specifically coded messages with more-or-less symbolic meanings. It's as if you discovered radio-frequency FM on a channel you thought was only carrying unencoded audio. So don't be too sure of those assumptions. --JoSH]