Will Millenials die before they invent human mind-uploading?

[Lago PARANOIA]

Discuss. It would suck quite a bit to be the people who died before total and complete immortality, muahaha.

[Foxwarrior]

Mind-uploading is a real mess and probably impossible because computer structure and brain structure are not very similar at all.

A facsimile of mind-uploading where the robot copy is only good enough to fool the now-dissected person's peers and family members should be doable in a couple of decades, though. So probably some of them will be bothering you for centuries to come.

Mind-uploading where the brain structure is stored digitally, but in order to run it it has to be bioprinted with fresh gray matter seems doable enough too.

[Username17]

Yeah, I don't think human brains can be downloaded into a non-degrading medium. The slow degradation of our substantia nigra is a part of how we think and who we are. If you replaced it with a semi-permanent crystal, we wouldn't be us, we'd be different people.

Now I fully believe that we'll eventually have the ability to create much longer lived solid state brains that are at least as "intelligent" as humans. And eventually we'll make those instead of fucking-based children. But no living humans are going to transform into those things. Everyone born as a human will die as a human, and the first near-immortals will have human parents rather than human caterpillar states.

-Username17

[OgreBattle]

I think there was a Den thread about differnet grades of AI and how if you made a human-like AI it wouldn't be something you could copy/paste.

[DSMatticus]

I think there was a Den thread about differnet grades of AI and how if you made a human-like AI it wouldn't be something you could copy/paste.

Intelligence is just data being interpreted by the process of execution. If the medium supports copy/paste (i.e. electronic data), then you certainly could. If the medium does not support copy/paste (i.e. the actual gray matter in your skull), then you certainly couldn't.

I suppose it's possible that the human brain achieves its amazing efficiency by blurring the line between hardware and software, and the only way to create an even remotely efficient AI would be to do something similar (as opposed to executing arbitrary brains on universal brain emulating hardware), at which point printing a brain would be less about ctrl+v and more about having a fancy 3d printer, but again: that's limitations of the medium you're working in, not the nature of intelligence.

[Dean]

No, we're centuries away from achieving such a thing if we can ever do it at all. People get really excited when you talk about computer technology and assume too much because they ignore the necessities of human research. Computer power can double all it wants you still need a guy to map how the brain works before you create a computerized replica. We are still largely clueless about how the brain works and there are no groundbreaking answers anywhere in the pipes. If any of us live another 50 years we'll probably see human facimiles good enough to fool other people and probably active computer systems that can learn in some useful way but actually going from grey matter to ram requires medical research on the brain and that shit is slow as fuck.

[DSMatticus]

I would not say centuries, mostly because projections about technology that far in the future are virtually guaranteed to be bullshit.

On the side of optimism, I want to point out that we have made artificial hippocampi for mice - an electronic device that takes as input neural signals and produces as output neural signals. That is fucking amazing. Progress is being made in this field. It's not easy or fast, but we're doing it, and interest in funding it seems to be on the rise. There's a lot of good reasons to be optimistic about the next half-century of neuroscience and artificial intelligence. Edit: not to the point of immortalizing brains, mind.

[radthemad4]

Okay so mind uploading isn't happening anytime soon. Any potential alternatives to immortality that we might be able to afford in our lifetimes?

[Ancient History]

Or you could be remembered for your deeds.

[nockermensch]

Okay so mind uploading isn't happening anytime soon. Any potential alternatives to immortality that we might be able to afford in our lifetimes?

We'll probably see some kind of drug or treatment that can extend telomeres during our current lifetimes. This, coupled with advances in stem-cell research, should allow the very rich to stop aging and their brain's degradation, and replace any failing organ, so extending their lives indefinitely.

No idea of how this would work on practice, but right now, immortality, if it comes, looks more probable via the bio-engineering route.

[Pseudo Stupidity]

I'm not sure about mind uploading, though quantum computing might allow for some brain-like shenanigans. Google bought a D-Wave quantum computer for machine learning about a year ago.

As far as "you" living forever goes, we can already 3d print living liver tissue. It's probably a matter of time before we can 3d print entire organs. Cancer of the X? Fucking cut it out and print a new one. That's not living forever, but it's certainly going to extend lifespans by a ton for people who can afford it.

[nockermensch]

I'm not sure about mind uploading, though quantum computing might allow for some brain-like shenanigans. Google bought a D-Wave quantum computer for machine learning about a year ago.

As far as "you" living forever goes, we can already 3d print living liver tissue. It's probably a matter of time before we can 3d print entire organs. Cancer of the X? Fucking cut it out and print a new one. That's not living forever, but it's certainly going to extend lifespans by a ton for people who can afford it.

But without some telomere extending trick, the cells that comprise most of your body will simply stop replicating after some point, even if your organs are in perfect health. The last time I read about this, telomeres seem to be work as hard-coded limiters for our life-spans. The good news is that we're already hacking with them.

[Username17]

There are certainly some forms of brain degradation that you can postpone, but the primary issue is that there are some forms of brain degradation that you can't reverse. If you live long enough, you will get Parkinson's and die. That is a thing that will happen, because long distance tracts in the brain cannot regenerate.

And if you turn off the astrocytes' suppression of new axon formation to allow those tracts to regenerate, it basically formats your brain as axons stab in every which way and you die a horrible death. Like, really horrible. I wouldn't wish that kind of death on anyone.

I could imagine some sort of nanosurgery that suppresses the brain's resistance to new growth in very small areas around guided microwires or some shit - but we're pretty fucking far into science fiction dream worlds at this point and I actually think we'll be able to raise an entire generation of designed superhuman cyborgs to adulthood before we have that working.

-Username17

[RobbyPants]

And if you turn off the astrocytes' suppression of new axon formation to allow those tracts to regenerate, it basically formats your brain as axons stab in every which way and you die a horrible death. Like, really horrible. I wouldn't wish that kind of death on anyone.

I have no idea what this means. Is this sort of like how that gene that stops us from regenerating also inhibits cancer growth?

[Mistborn]

There are certainly some forms of brain degradation that you can postpone, but the primary issue is that there are some forms of brain degradation that you can't reverse. If you live long enough, you will get Parkinson's and die. That is a thing that will happen, because long distance tracts in the brain cannot regenerate.

I always assumed that any form of human immortality would be ship of theseus style. Replacing parts of the brain with something synthetic over the course of decades.

[Grek]

As people have posted, the answer is "Almost certainly."

[Username17]

And if you turn off the astrocytes' suppression of new axon formation to allow those tracts to regenerate, it basically formats your brain as axons stab in every which way and you die a horrible death. Like, really horrible. I wouldn't wish that kind of death on anyone.

I have no idea what this means. Is this sort of like how that gene that stops us from regenerating also inhibits cancer growth?

While it looks similar under a microscope, brain tissue isn't liver tissue. The liver is composed of basically a pile of hepatocytes in a honeycomb. As long as you get them draining bile in the right direction, one pile of liver cells is pretty much as good as another. However, neurons have exceedingly specific ways of connecting to each other, and that's really important to their function.

Neurons connect to each other via axons: processes of the cell that release chemical signals from the end and use electrical signals to communicate between the main cell body and the end. Some of these axons are really long: your sciatic nerve is a bundle of axons that runs from the base of your spinal cord to the end of your foot. So depending on how tall you are, those axons can be over a meter long - all as a very thin projection of a single cell (well, a braided rope of cells that all have very thin processes that run that distance, but you get what I mean). When an axon is cut, it has the capacity to regrow. It grows at about 1 mm per day, meaning that You can cut the sciatic nerve and regain control of your foot a couple years later (subject to the pathway getting blocked by scar tissue or something).

Where this all gets fucked is in the central nervous system. You have these axonal pathways that transmit information all over the brain. Bundles of these axons are called "tracts." Some of these tracts are relatively simple (vision gets pumped from the eye to the thalamus in a roundabout but conceptually simple fashion), and others are really complex. For example, dopamine is produced in the ventral tegmentum and substantia nigra pars compacta, but you have at least five types of dopamine receptor expressed in different parts of the brain where it might be transported to via those axonal tracts.

We don't actually know enough about the central nervous system tracts to draw out where they should be going, but that doesn't really matter because after you reach a certain stage of development your brain shuts down the ability of axons to grow. Once the neurons that project the axons that make up the longer tracts die, it doesn't actually matter whether you can grow replacement neurons, because the axons aren't allowed to grow long enough to remake the old pathways.

Now the whole "none shall pass" thing for axon growth is of course a chemical signal (produced by astrocytes that exist in the central nervous system and not the peripheral nervous system, which is why the nerves in your foot can regrow while the tracts in your brain can't). And like any chemical signal, we can turn it off by any of a number of methods. But it turns out that if you do turn off that signal, the brain makes itself a bunch of crosslinked axonal connections and the whole thing shorts out. It's not actually cancer, but it might as well be. It's like getting a lethal form of advanced autism.

So while we can certainly extend life by decades by swapping out hearts and lungs and livers and spleens with new shiny ones grown in genetically modified pigs or something - it's not at all obvious that it is even possible to Ship of Theseus the brain. There are probably bits in there that we'll never figure out a way to replace in a living human without killing that human. Actual, or even near-immortality is something which will only be available to people who have been designed from the beginning to be repairable without reformatting. Which means none of us. Also none of our children. But possibly some of our great grand children.

-Username17

[radthemad4]

And like any chemical signal, we can turn it off by any of a number of methods. But it turns out that if you do turn off that signal, the brain makes itself a bunch of crosslinked axonal connections and the whole thing shorts out.

Would it theoretically be possibly to turn it off very briefly, manually trim unwanted growths before they can cause any damage(possibly via nanobots or something), turn it on again and repeat until the desired growth is achieved?

[Nachtigallerator]

I was going to say something to that effect, but Frank put it much more succinctly, so I'll just add stuff:

- For the same reason (adult brain cells do not form entirely new connections and with a few exceptions do not regenerate), it's not possible to transplant central nervous system cells - which includes brain parts and the spinal cord.

- The perhaps most problematic part of cancerous growths is that they spread metastases. In fact, most of the tumors that are considered very deadly are considered such because they metastasize early and to important organs, so operations have to happen early or are basically pointless. If you can catch lung cancer when it's just a local growth, you can indeed cut it out and reasonably hope to be done with it - but that rarely happens because it's not very noticeable, and it will spread all over the patient.

So even with unlimited autologous transplants, there's a limit to the amount of organs you can replace in a single person before the operations take a toll or go wrong. Tumors can have all kinds of side effects that are hard to treat - And of course, cancer can always spread to the brain. Melanomas are notorious for that.

[Username17]

And like any chemical signal, we can turn it off by any of a number of methods. But it turns out that if you do turn off that signal, the brain makes itself a bunch of crosslinked axonal connections and the whole thing shorts out.

Would it theoretically be possibly to turn it off very briefly, manually trim unwanted growths before they can cause any damage(possibly via nanobots or something), turn it on again and repeat until the desired growth is achieved?

The desired tracts are much longer than the kinds of crosslinking that will make you smell the color blue until you lose control of your bowels and stop breathing, so I'm guessing no.

I could imagine a scenario in which we ran a very tiny tube along the tract's pathway and then suppressed the growth prevention in a local fashion somehow - essentially allowing axons to grow along the preferred tract route and not other places. While we're wishing for ponies, we could have some system that speeds the growth of axons, so you aren't waiting for six months with guidewire stuck in your brain for the tract to regrow.

But even if such a system can keep people from dying of Parkinson's, it seems pretty unlikely that every brain insult would be amenable to techniques like that. It's one thing to lay down new sensory or motor tracts - those are fairly standardized; but when you're laying down replacement cognitive tracts or memory associations, you're basically just making a new person.

-Username17

[fectin]

At the point at which we could usefully do the thing Frank is imagining, we already have such a detailed model of how your individual brain works that for all practical purposes we have already Morovac transferred you.

Frank also claims that sensory and motor tracts are fairly standardized. This is a lie. They might be, and it's so reasonable to believe so that it gets into textbooks. But there is zeero research backing it up. We can't even show that on peripheral nerves, where it all acts like wire bundles.

And just for giggles, everything we know about how brains work now seems to get disproved every few years. Nachtigallerator's claim that we don't form new connections, for example, is a whole 2.5 months out of date (http://news.wustl.edu/news/Pages/26323.aspx). That's not to say he's wrong so much as it is to point out that brains, as an anthropomorphized group, seem to delight in fucking with researchers.

[Nachtigallerator]

Sorry - I was not being specific enough. I simplified "tract" to "connection", whereas a "tract" is a large axon highway that anatomists can actually see (in a dead brain) with proper staining, and that radiologists can, to a limited degree, actually demonstrate in a living brain, too. I assume that this is also what Frank was talking about when he said tracts were standardized - and sorry, but that is right. Anatomy textbooks are aware of variations in the anatomy of the superficial veins in the arm, and we have cut and stained enough brains to make statistical statements about normal human anatomy. And in normal anatomy, the tracts look quite similar between people, which is very good for neurosurgeons. If you google "neural tract variation" what you will see on page one are reports about size variation, not variant anatomy.

It is clear that something morphological is happening to generate memories and learning, and that somthing involves connections between cells because this is how the brain does anything at all. Although I want to point out that while this study gives reasonable grounds to assume that there are new connections, the article about it nowhere specifies how exactly these are formed or even give an anatomical description. It most certainly is not about demonstrating the formation of a new, macroscopically visible tract in the sense of the corticospinal tract.

It is beside my point anyway: At the same time, what clearly isn't happening is that brain cells from the frontal cortex send a tract to the optic nerve and start being the primary area for visual processing because the occipital cortex is broken. That degree of flexibility does not exist for brain cells.

[RadiantPhoenix]

What if we invent a time machine and meet friendly people from the far future, who have millenia of extra knowledge about how the human brain works?

If that happens, is it halfway plausible?

[fectin]

I assume that this is also what Frank was talking about when he said tracts were standardized - and sorry, but that is right.

The tracts are like a cable run: it's easy to point and say, "all the cables go through here." That is what you're talking about. Manipulating that in anything other than a gross effects kind of way requires knowing about the individual strands inside the individual cables. We don't even know where the cables go.


Here:

This is a feline sciatic (stained; in agar; mounted on a vibrotome).
A) this is the nerve overall. It's roughly equivalent to a tract.
B) This is a fascicle. They noticeably move around even within a centimeter or so, and we can't repeatably select fascicles to interact with. FINE and SPINE electrodes attempt to address this, but we don't even know if there even is a standard anatomy for fascicles.
C) These are axons (Tiny purple dots, here). They don't show up on this picture very well, because I couldn't get the camera in exactly the same focus as the eyepiece. They are the actual nerves that actually carry the signals.

[nockermensch]

When I tried to see this image I got this error:

Warning: Unsafe Website Blocked!
wiki.fectin.com

This website has been blocked temporarily because of the following reason(s):
Phishing

I use the dns from comodo antivirus as an additional protection against malware. This can be a false alarm, but it wouldn't hurt to check what happened.