X-Message-Number: 12635
Date: Mon, 25 Oct 1999 16:33:45 -0400
From: Mike Darwin <>
Subject: Memory and priorities

John,

You write that: 


>They found that LTP spreads out over a large area so you have lots of
copies
>of the same thing. Neural Net expert Terrence Sejnowski said at the time
>"Instead of thinking of a synapse as representing a piece of information
you
>can now begin thinking of a population of potentiated synapses acting
together".
>Could the same thing be true for Long Term Depression?



I think where "we" differ is not on what might be possible, because I think
that almost anything that acts within the broad confines of what we
CURRENTLY understand as physical law is possible. What you are really
asking me is more like: "Mike, don't you think it is PROBABLE (to the point
worth gambling people's lives on) that such and such is so....?" 

There is a longing in such interpretations as "yours" and many others in
cryonics which I find really corrosive. It should be FEAR or ANXIETY about
that which UNKNOWN rather than relief at that which is known which sets the
agenda. 

I'm a biologist and a medical guy, not a computer weenie or an engineer. As
a consequence, I am very leery of favorable, best case scenarios as a basis
for planning and action *where there is any other reasonable alternative.*
Now, if you're dying of cancer in a few days, you have every right to take
the most optimistic outlook about every aspect of cryonics possible.
Ironically, the sad fact is that when cryonicists REALLY start to die they
are often terrified and morbidly pre-occupied with the very uncertainties
they were happy to ignore or minimize when it was SOMEONE ELSE S more
immediate problem.

Frankly, I've never been a big fan of the *robust* redundancy theory of
memory storage. Sure, there's redundancy, there has to be.  But, the hard
thing for people to understand is that redundancy, to be meaningful, has to
follow one of two paths which I can best communicate as follows:

1) If you are using redundant media to store information that are
vulnerable to the same kind of attack you can only protect your information
by methods that isolate the redundant information from the same kind of
insult. Example: having 5 copies of  Walt Whitman s LEAVES OF GRASS will do
you no good if they are all in the same room which burns to ash. You have
to put a copy in a fire-resistant safe, or locate one off-site, and so on,
if you want to use the *same media.*

2) The safest course of action is to use different media with different
weaknesses and make multiple redundant copies separated, if possible, in
time and space from each other. Greg Benford has written a great book on
this problem called DEEP TIME which he was kind enough to send me a
complementary copy of. I never thanked him, so, if you re reading Greg, it
was a great book because it was from the heart and it made me *think* about
problems I already thought I d given plenty of thought too. (BTW it is one
of the best documented examples of the insane waste of your tax dollars
that I can think of.)

My concern is NOT with LTP, LTD or any other specific method of memory
encoding beyond their characteristics of density and infer-ability as they
relate to current or anticipated human cryopreservation protocols. Rather,
it is based on a DEEP EVOLUTIONARY OBSERVATION that there is no "incentive"
for the brain to store memories (except by radically different classes of
function: i.e., declarative versus procedural memory) in any but the same
way, redundantly or not. Thus, its like having five copies of a book in a
room that burns to ash: great if the fire (damage) is local and the books
are distributed all over the room. 

I have grave concerns that just the act of CPA loading and unloading may
perturb membrane structure enough to irretrievably destroy higher-level
declarative memories. I've seen too many humans and animals with diffuse
brain injuries wake up and react with horror (in the case of  the human
I ve seen) when her husband tries to kiss them: "Who the hell do you think
you are? I don't KNOW you! I've never seen you before in my life." I've
actually seen this a few times. Such people rarely recover any quality
declarative memories: usually just bits and pieces, or nothing at all. 

This kind of experience "told" me a long time ago that such memory
mechanisms may be  fragile  in the sense of being susceptible to
destruction by a *global* rather than regional insult. Such injury is
virtually never survivable in nature, since such patients require severe
osmotic dehydration of the brain, ventricular venting of CSF, and prolonged
mechanical ventilator support.  Nature has no incentive to develop robust
different storage mechanisms as far as I can see.

Finally, I think that while "memories" of an event are stored in various
areas of the brain, they are stored in significant measure by TYPE. For
instance, I've known a cortically blinded women who can remember what was
said at her wedding, who was there, and so on, but NOT ONE SINGLE PIECE OF
VISUAL MEMORY: what anyone looked like, a scene from the wedding, or the
color of certain objects or things that were not cross-encoded verbally.
For instance she remembers the color of their bridesmaids dresses, but only
because it was a heated issue that was a subject of much VERBAL discussion:
she has no visual memory of the event. Such people are deeply fractured and
unhappy if they are fully aware of their situation. If they were highly
visual people, they are usually hopelessly depressed and, if capable, often
commit suicide. 

My main message is that there is no reason for complacency. We have the
tools to solve these problems or at least answer the questions at  hand. If
we fail to do so, and spend our time debating and playing games, it is
quite possibly, tragedy of terrible proportions. In any event that is a
chance we do not need to take and, considering the magnitude of the harm
and what is at risk, it is a chance we should NOT take.

Mike Darwin

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=12635