X-Message-Number: 22259
Date: Sun, 27 Jul 2003 10:03:32 -0400
From: Thomas Donaldson <>
Subject: CryoNet #21722 - #21727

For Francois, thought I should apologize for the lateness of this
reply:

You had a scenario for turning a human being into a computer. It
fails not just for possible philosophical reasons, but because
it seems to have an underlying assumption that at least our 
neurons work like computers.

First, signals are sent through structures (and lines) called
axons, and received through other distinct structures called 
dendrites. The connection between them is a synapse. Second, memory
in our brain (I refer to true long term memory, not the changes
at a synapse which occur at an early stage of learning and 
allow the longer stages to come into force) involves the growth
of new connections between neurons. Even as single bits it does
not exist in any single neuron, but in its connections to many
others. Third, at least 2 brain areas and some scientists think
ALL brain areas produce NEW neurons, which form new connections 
and all the rest. The dentate gyrus, in our hippocampus, lies
deep inside our brain; the other unquestioned area which makes
new neurons is the lining of our ventricles (areas within our
brain which are filled with a fluid very like spinal fluid). 
Fourth, neural signals are not single bits. It turns out that
the most efficient way of sending signals between neurons 
consists of repeated blips at a fixed rate. This is the most
efficient because (remember?) many neurons have thousands of  
connections and many of them may all be working at once;
some way to distinguish one message from another is needed.

Given that the number of neurons to which one neuron might
someday connect is very large, it won't work to have wires
between all of them at once, and rather than grow connections
activate already existing ones. Your nanomachine which 
introduces new neurons would also have to be constantly
active to deal with growth of new neurons... and their
multiple future connections. In terms of imitating brain
activity, that nanomachine would need to be built as a PART
of the brain rather than something that comes in from
outside and then goes away, leaving a human with a 
computer brain.

At this point it becomes unclear whether a system which
grows new connections and new neurons in response to its
own reactions (inner and outer) might not look a lot like
the biological system it was designed to replace.

            Best wishes and long long life to all,

                 Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=22259