X-Message-Number: 795
Date: Sat, 2 May 92 22:39:31 PDT
From: ghsvax! (Hal Finney)
Subject: CRYONICS: neurosuspension

I share some of the concerns expressed about neurosuspension.

First, I don't think the DNA in your head contains enough information to
allow an exact duplicate of your original body to be constructed, even with
nanotechnology.  The DNA contains a "program" which, starting with a
fertilized egg in a maternal womb, took nine months to build the body you
had when you were born.  But a lot happened in those nine months.  You hooked
up with the maternal blood system; you were exposed to thumps, bumps,
different orientations; you were exposed to nutrients and chemicals from your
mother's blood in different amounts at different times.  No record
exists of any of that information, and it has to have influenced the
details of your body's development.  (Not to mention all the things
that influenced your body's growth the rest of your life.)

Identical twins start with identical DNA, and have very similar
maternal environments, but they are not cell-for-cell identical.
At best, DNA would allow the construction of a body which was as
similar to yours as twin baby's bodies are.  (And probably that ideal
can't even be approximated that closely; it would be more likely as
similar as are the bodies of adult twins.  Often they don't even look
like siblings at first glance.)

I find especially bizarre Thomas Donaldson's suggestion that it would
be at all easy or natural to grow a new body from the neck down.  The DNA
contains no such program.  Presumably as this body is growing down
little proto-organs, little hearts and stomachs and lungs, are growing.
Do they maintain topological connections with the adult-sized head
this whole time, or do they hook up later?  How are you going to find
the information in the DNA that tells what the mature, adult-sized heart is
supposed to look like?  The DNA program only tells how to make a heart
by starting from an ovum in a womb.  It doesn't tell how to grow one
out of a neck.

Assuming we did make a body by growing it from a cell (or simulating
that process on a computer and then using nanotech to build an
identical final product), we then have to attach the head.  But I
don't think the neural networks are going to be completely compatible.
The spinal cord is packed thick with neural axons, and probably the new
body won't even have the same number and arrangement of neural processes
as the head.  The mapping will be between two topologically
incompatible networks, like cutting Paris and New York in half and
trying to butt them together and match up all their streets.

Even if this can be done, I don't think the information about exactly
what neural axon is supposed to go where will be obtainable by passive
means.  One of the characteristics of neural networks is that they
learn in such a way that you can't really tell by just looking at the
network exactly what they've learned.  In practice, you basically have
to run the neural network and see what it does for different inputs to
find out what information its connections encode.  (When I say "run",
I include both computer simulations and/or biological activity.)

In other words, given a head, and a complete map of all the neural
connections in it, and how they lead down to the spinal cord, that
probably won't be enough for you to figure out that this particular
spinal cord neuron was supposed to go exactly to such-and-such a spot
on a particular muscle in your left little finger.  The brain isn't
labelled like that.

I believe the only way to get this information is to run the neural
net with a variety of simulated inputs.  Running enough of these "can
you feel anything when I do this" and "try to wiggle your big toe"
tests will eventually provide the information needed.  But whether
this is done on a biological brain or a computer model of one, it
could amount to subjective years or decades of boring and possibly
uncomfortable experience before I could be fully re-built.

In short, although I think neurosuspension could work, I feel that the
lost information about the physical form of a body which is compatible
with the head can be recovered only through a laborious process that
could be quite unpleasant for the head itself.

Hal Finney
ghsvax!			...!uunet!ghsvax!hal

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=795