X-Message-Number: 5769
From: Anders Sandberg <>
Newsgroups: uk.legal,sci.cryonics,sci.life-extension
Subject: Re: Death (was Donaldson MR and Miss Hindley)
Date: Mon, 19 Feb 1996 16:16:01 +0100
Message-ID: <>

References: <> 
<>

On Sun, 18 Feb 1996, Marshall Rice wrote:

> In article <>
>             "Marvin Minsky" writes:
> 
> > It is possible that all one needs to recover are the approximate
> > geometries of the previously existing synaptic connections, and some
> > coarse characterizations of their propert ies.  
> 
> Unfortunately not. We would need to duplicate the precise properties of each 
> synapse, i.e. at what level of stimulation it was triggered and for how long.

This is one of the big unsolved questions today: what levels of activity are
important? I have so far never seen any hard data on this, and the question
seems to be unsettled. On one hand, synaptic strengths do play a role, but
are they qualitatively or quantitatively important on larger scales? If their
exact strengths are important, then *very* minor changes in chemistry would
have disastrous effects on our thinking. It is worth noting that a specific
type of cell seems to have identical types of synapses (with minor individual
variations). If it really turns out that Long Term Potentiation is based on
local-area diffusion of NO, then memory is distributed between thousands of
synapses and the loss or misrepresentation of a few would not matter. In the
same way local neural networks could be very resilent to minor disruptions. 

> The problem lies in re-establishing the neural pathways, not the neurons 
> and synapses themselves. The fact that there are more potential neural 
> pathways in a human brain than atoms in the solar system, gives some idea
> of the scale of the problem.

But this assumes nothing is known about these pathways, and that they 
can go just about anywhere - most of those possible pathways would not 
correspond to a viable (or human) brain. The actual number of 
possibilities is still huge, but can further be constrained by additional 
data like the remains of the neurons and future knowledge of the rules of 
neuronal growth (we are making some impressive progress in this area 
today). 

For example, if I find a broken axon-end I would look for other broken
axons-ends in the vicinity, taking advantage of what I have observed about
the local area (freezing crack directions, microtubule patterns in the axon
end, etc. ), if I found a loose synapse I would start looking for a broken
dendritic spine and so on. Gradually I could refine a local map of what ought
to be connected to what using maximum likelihood estimation and other
techniques. 

See http://merkle.com/merkleDir/cryptoCryo.html for some further ideas 
about this. 

> Those who are about to jump in and cry
> "future technology" should think about that for a few seconds.....

People have thought quite a bit about it. 

> I cannot conceive of any way in which neural pathways could be restored 
> other than by (1) reconstructing each neuron and synapse down to the 
> molecular level or (2) re-accomodating the synapses. 
> 
> In the case of the latter (and quite probably the former), although you may 

> be able to duplicate the personality, emotions and memories of an individual,
> in no sense would they be the same person. The consciousness of the original
> individual would have been lost, just as surely as if they had gone to feed
> the worms.

This of course assumes that consciousness is not based in the brains of 
people (or more specifically, in the structure of their brains). Cryonics 
seems to be one of the few ways this could be tested experimentally. 

> As to experiments, we have a considerable amount of data from victims of
> injury and disease.

There are also opposing evidence to your above assumption, such as 
drowning victims who have been revived after a considerable time in cold
water. Despite the fact that they had no brain activity for a while, they 
recovered (in many cases without lasting damage). As for neuronal 
restoration, time will tell; there is promising work being done about 
implanting fetal or immortalized cells, neurotrophic substances and 
possible electronic implants. 

 -----------------------------------------------------------------------
 Anders Sandberg                                      Towards Ascension!
          http://www.nada.kth.se/~nv91-asa/main.html
 GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=5769