X-Message-Number: 4394
From:  (Thomas Donaldson)
Subject: NNs and our brain (and brains of salamanders)
Date: Fri, 12 May 1995 12:59:32 -0700 (PDT)

Mr. Eugen Leitl recently made a post commenting on what I had to say both
about brains and about the powers or lack of powers of animals such as 
salamanders to recover memories after brain damage.

While I've been interested in NN as computing devices for some time, too, I was
speaking in terms of the neurobiology of the subject. In this case, the 
"strength" of connections which is so important to the operation of a neural 
net SEEMS to be implemented by having multiple connections between the same 
two neurons. That is why connectivity is very important. If anyone wants a
reasonable summary of these ideas as applied to real brains, they might read
Steven Rose's book THE MAKING OF MEMORY. One of the very simplest observations
and one known for a long time to support this is the fact that rats that 
learn a lot have a far more extensive neuropil than those that do not.

In this sense it becomes important to know about connectivity because that is
the way our brains implement "strength". I don't wish to put this point 
forward as certain --- our understanding of how actual brains work is still
too little -- but that is what I meant.

And naturally, if we look at animals which can regrow their brain connections,
we would (on this theory) want to know whether they make not just one but
many more connections when the link between two neurons was formerly one 
with many connections.

While this notion of brain connectivity IS just a theory, one set of studies I
have not yet seen is that of working out the behavior of an actual neural net
which used this connectivity method to operate. It's far from obvious to me
that it will behave exactly the same as other neural nets, though I think it
can be classified as one. To the degree that such a NN behaves differently, it
cannot be easily embodied (?) in a DIFFERENT kind of NN. 

And I don't mean to say here that using neural nets of standard kinds to work
out how our brains work is useless at all. They will give us good ideas about 
that. But these remain good ideas until and unless they are verified by the 
actual operation of a real brain. TJ Sejnowski, for instance, who has done a 
lot of work of this kind, is quite open about the fact that on the node level

ie the level of individual neurons and how they work -- his neural nets 
probablydon't match our brains very well.

As for storing people on tapes, yes, that would work fine. Ideally, of course,
you want some storage medium that deteriorates very slowly on a scale of 
centuries. If and when we can read people off into a computer, a lot of work
will be done on finding a medium which does that --- right now I doubt it is
true for anything but writing on paper or engraving on stone.

Finally, if Mr. Leibl believes that reading out the structure and connectivity
of a brain despite freezing damage is now possible, I would like to know much
more about his specific implementation of this idea. If he does not have one,
then he should label his comment as a theoretical speculation. 

			Long long life,

				Thomas Donaldson


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=4394