[ExI] [Extropolis] Crosspost

[Keith Henson]

On Sat, Apr 19, 2025 at 5:03 AM John Clark <johnkclark at gmail.com> wrote:
>
> On Fri, Apr 18, 2025 at 10:09 PM Keith Henson <hkeithhenson at gmail.com> wrote:
>
>>> >> There's no disputing matters of taste. As for me I'd prefer a destructive scanning that didn't corrupt information over a non-destructive scanning that did.
>>
>> > You might convince me that destructive scanning preserves information
>> better than non-destructive scan.  Can you make such a case?
>
> I can make a moderately strong case that destructive brain scaning has already occurred a few times, but nothing even close to a non-destructive scan exists today. Of course both will improve in the future but destructive scanning has a significant head start.  Today several human brains have been chemically fixed with ASC , sliced with a diamond saw into many very thin slices, the slices were photographed with a high power microscope, and then computers were used to analyze the photographs and trace out the neural connections.

I am a long way from thinking that the wiring diagram is enough to get
a person back from suspension.  From what we know, synaptic weight is
essential to memory.  I think without this data, which as far as I
know is not visible, you get back a blank brain with no memory.  I
don't know about anyone else, but to me, a no-memory on revival
cryonic suspension seems pointless.  Your case would be much improved
if you could show that memory could be recovered from a scanned brain
slice.

> The last two steps were the slowest and the most expensive, and those are also the two steps that are likely to improve the fastest in the coming years; and an organization such as ALCOR need not worry about those two steps because that is a problem for the future.
>
> Fun fact: one of the brains that were treated in this way was that of a convicted murderer who had been executed by lethal injection.
>
>>>> >>>
 There is no reason I can see why all the structures in a brain could
not be mapped out by infiltrating it with nanomachines.
>>>
>>> >> That might be possible but it would be slower, harder and more expensive than a destructive scan.
>>
>> > I am curious why you would be concerned about slower.
>
> If you're going to infuse a living human brain with nanomachines to record how things are wired up and expect the brain to continue to function normally then you're severely limited in the number of nanomachines you could put into that brain,

Given the relative sizes of nerve cells and nanomachines, I doubt
that's much of a limit.  But it is something we can put numbers on.
Brains function fine after swelling 5-10 percent.  How many machines
is there room for?  A bigger problem might be getting the connection
and synaptic weights out of the brain.  I have given a little thought
to this.  The reason you want to go slowly is to keep the waste heat
down.

[From the Clinic Seed story]

She was mildly distressed that she now had to voice talk to Suskulan,
who appeared as a projection, instead of "talking" directly to his
spirit in the spirit world she had inhabited.  Then she realized from
her new knowledge there was a way she could if she took a bit of the
clinic with her.  However, there wasn't much time to before her
parents came.

"Can I come back to visit even if I am not hurt?" she asked.

"Yes.  Anytime I don't have another patient."

"May I take the clinic's interface with me?"

"There is nothing so addictive . . ." thought Suskulan.

"You may."  Part of the cloud of nanomachines that had just left
Zaba's brain returned as a momentary haze.  Since they retained their
memory of where they had been it was a matter of a few minutes before
the machines reestablished their monitoring posts in Zaba's brain.

"I missed not being able to talk to you in the spirit world."   Zaba
said without voicing.  A wire frame image in Zaba's visual cortex
overlaid the physical projected image of Suskulan.

"Spirit talk does not reach as far as your garden."  Suskulan warned her.

Zaba lay down on the repair table that was now at the bottom of the
elevator shaft.   The elevator lifted it into its place in the clinic.
  Zaba was treated to seeing the rapidly thinning utility fog image of
her body that had comforted her family for the last ten days before
she merged into her image.

The nanomachine haze that had fogged her image and now her real body
withdrew into the low table.  She greeted her family as they came into
the clinic and in voice talk said goodbye to the image of an old man
Suskulan was projecting.  Then they stepped through the clinic's
keyhole door to where the other members of the tata were waiting for a
joyous celebration of the healing of Zaba.

Suskulan sent off a strictly factual report.  There were no replies
this time, but perhaps that was due to the high report traffic.

Her family had visited every day, but they were still delighted and
relieved that Zaba was back with no visible effects from being shot.
Her parents had been worried that her value as a bride might have been
damaged, but none of the tata seemed to be concerned, only very proud
of the growing powers of their clinic Suskulan.  (The elders had long
since wildly inflated the value of the fetish they had traded for the
clinic seed.)

Zaba had been warned not to flaunt her new knowledge to adults and
with Suskulan's help had built temporary inhibitions into her mental
processes.  She was under no such injunction toward the other
children, though.  They were absolutely fascinated and wanted the
ability to talk to Suskulan in the spirit world as well.  In spirit
world talk Zaba asked Suskulan if he would give the others an
"interface" like she had.

"Yes, though not in one day like I did with you.  It takes several
days to a week for an interface to establish itself unless you are
very cold."

[end quote]

> and that limitation is going to slow things down. I might add that if a mechanic is trying to change the spark plugs on an engine his task becomes much slower and more difficult if he is not allowed to turn that engine off while he works; it also makes it more likely that he will make an error.

This is not a problem since low temperature will shut down a
biological brain just fine.

With cryonics patients at LN2, you are starting with a shut-down brain.

>> > it's not like humans would be doing this, as long as it is automated, who cares about harder or expensive?
>
>
>  We know for a fact that ASC preserves the synaptic neural connections in the brain better than the procedure ALCOR currently uses because today we can detect those connections if ASC is used but cannot do so with ALCOR's procedure. We have some reason to be hopeful that ALCOR's procedure also preserves that information and it's just scrambled up more, but we do NOT know that for a fact. And why make things more difficult for future technology to bring us back if that difficulty can be avoided?
>
Do you have any pointers to ASC preserving synaptic information?  That
would be very interesting, essentially reading out memory.  As far as
I know (and I may be out of date) they can see synaptic nodes, but
pictures do not disclose the weighting of a node.

> This is what ALCOR had to say about ASC back in 2018 and as far as I know they haven't said anything about it since:
>
> "A new cryobiological and neurobiological technique, aldehyde-stabilized cryopreservation (ASC)  provides strong proof that brains can be preserved well enough at cryogenic temperatures for neural connectivity (the connectome) to be completely visualized. [...] Current brain vitrification methods without fixation lead to dehydration. Dehydration has effects on tissue contrast that make it difficult to see whether the connectome is preserved or not with electron microscopy. That does not mean that dehydration is especially damaging, nor that fixation with toxic aldehyde does less damage."
>
> ALCOR's position on brains preservation
>
> I would maintain that the last sentence in the above is factually incorrect. ASC DOES cause less damage than ALCOR's current method. That's why we are able to trace the neural connections with today's technology with one method but not with the other. The damage caused by ALCOR's method may not be irreplaceable, the information may just be scrambled more than it is with ASC and require Mr. Jupiter Brain to jump through more hoops to recover it, but maybe not, so why take the chance?
>
>>> >> we are both ALCOR clients and if we're lucky enough to be revived I don't think we will have much say about how it was done.
>>
>> > You can write specifications into your contract.  One Alcor patient
>> is/was blind.  He specified that he is not to be revived until the
>> procedure can give him sight. But you are essentially correct.
>
>
> That seems unnecessary, if the future people have the technology to repair a freeze damaged human brain they certainly have the technology to restore his sight.

It is not a matter of restoration, the patient was blind from birth.
He was one of those blinded by preme oxygen treatment.

>I wrote no specifications in my ALCOR contract because I thought it unlikely that anybody would pay attention to them and if they did they might turn out to be counterproductive because I have only a hazy understanding of what the post singularity world will be like. For example, if somebody wrote that they do not wish to come back as an upload and that request was honored I don't think Mr. Jupiter Brain would bring him back at all.

Possible.  However, I think repairing brains/bodies is on a par with
uploading.  I suspect that uploaded humans will find that state more
desirable than the physical state, leading to a population collapse.

Keith

> John K Clark   See what's on my new list at  Extropolis
> 288
>
>
>
>
>
>> >> > More cell damage occurs during the thawing process than the freezing process, and if ASC chemical fixation is used there is no brain shrinkage and the synaptic connection information is preserved; we know this because beautiful electron microscopic pictures have been taken of brain cells preserved in this way. Then the frozen brain could be disassembled from the outside in, one very thin layer at a time, and the information about where and how strong all the synaptic connections in that layer could be recorded, and then work could start on the next layer and you keep going until there is nothing left of the brain.  After all the information in all 10^14 synapses have been recorded that information is later translated into electronics and the uploading has been completed.
>> >> >
>> >> > OK OK I admit the above scenario may seem like a crazy fantasy but it should be remembered that, unlike perpetual motion or faster than light spaceships or traveling to the past, it does NOT need to invoke new science to become a reality, all it needs is improved engineering.
>> >> >
>> >> >> > I also doubt that ET beings create mega-structures or planet sized computers or brains.
>> >> >
>> >> >
>> >> > I too think ET is very unlikely.
>> >> >
>> >> >>
>> >> >> > If it does happen it might be in one out of a trillion galaxies. It will not be done by us. I suspect we will be off the Darwinian game table in the rather near future.
>> >> >
>> >> >
>> >> > I very much doubt biological humans will still be around a century from now and perhaps not a decade from now, but I have some hope that Mr. Jupiter Brain will have at least a little affection for us, after all He wouldn't exist except for us, so maybe He will give us access to a small (by His standards) server so that a few billion uploads can be run in a pleasant virtual world.
>>