Knowing someone with a spinal injury I can tell you that these constant innovations that won't reach the market for 20 years are kind of continuously devastating.
There is too much focus on the "cool" factor and not enough on quality of life and results for impacted patients. I don't know how to fix it.
I've mostly given up on commenting on this on HN. I have an incurable condition and it makes zero difference that these practices directly negatively impact my life and scare the hell out of me, such observations get down voted and attacked and treated like crazy talk.
I'm sorry for what your social contact is going through.
I have a spinal cord injury (well rather, an incomplete spinal cord injury) and knowing that every day will be like the last until I die has taken a toll that has at times almost led me to bring that event forward in time. Look after your nerves my peeps.
20 years is infinitely better than never. There are a lot of life-altering diseases whose future outlook is much bleaker (eg. little funding or research interest). But yea, having to wait so long definitely sucks for those currently suffering.
I'm not a medical person, can anyone who knows more than me explain exactly how cool this is? Fixing spinal cords, or just repairing simple nerve severance?
Too early to tell. Might make less invasive neurostimulation therapies, may or may not make them more effective (not all neurons can grow back or grow back to where they are supposed to be).
Even small nerve injuries can be life changing. Friend my motor cycle rat friends hit a bridge abutment. Severed the nerves in his right shoulder. He has no feeling in and cannot move his arm. So being able to fix that would be a big deal for people affected.
This is almost probably not what you want. Your nerves respond pretty quickly already and you can do things faster than you can think about them (e.g. catching a ball) so there's more to the computation than the "CPU".
I have read that the myelin sheath on your nerves is thicker at birth then as an adult. The theory (not mine!) is that as it thins the capacitance goes down and perhaps this helps keep the transmission time more constant as yu grow, so that what you have learned about walking, catching a ball, drinking hot coffee from a cup continues to work. Imperfectly -- kids are clumsy -- but enough that you can quickly adapt and not have to keep learning all over again.
So if as a child I learn to send a signal to my hand to move out to catch the ball, the signal takes X microseconds to travel from brain down my arm to my and and "clasp"
But as an adult / teen my arm is now a foot or more longer. So the signal would take x+y microseconds to get to my hand - but if the resistance of the nerves reduces it's signal is faster and back to x microseconds
Next level: link together a group of humans. I can't really imagine the implications of that, but I think of Vernor Vinge's "A fire upon the deep" and the group minds of the wolves there...
There's more than that to playing - a musician masters their own body in order to play the instrument with the required dexterity. There would need to be a translation layer between the commands that work on their body and the ones that'd work in yours (different finger lengths and muscular strength is the first thing that came to my mind).
I was rather thinking about instant communication and cooperation, instead of remote control. Like an entire orchestra tuned together - no more individual players but one orchestra superbrain. Or football team. Or of course army shudder
> To establish this digital bridge, we integrated two fully implanted systems that enable recording of cortical activity and stimulation of the lumbosacral spinal cord wirelessly and in real time (Fig. 1a).
I'm not sure it'd be a brilliant idea from an infosec point-of-view - the links themselves would need to be hardened in such way interfering with them would be impossible, or someone would be able to just blast the proper messages to the receivers to take over someone's body.
In fact, the conductor was half-joking. I'd probably go for fiber, even if it adds latency at the endpoints.
I don’t think it will ultimately help a diabetic. More worrisome then loss of feeling is loss of circulation. With High blood sugars you’re essentially get sugar crystals that build up on the capillary walls, which prevent blood circulation. Even if you cure Neuropathy will still have the issue of poor circulation.
But with neuropathy, which causes loss of feeling, pain… in the feet for example, you can ignore damage that combined with a loss of blood circulation capabilities can make things deteriorate very badly (necrosis, amputation).
If I understand correctly, this allows nerve-to-digital and digital-to-nerve conversions. This means that we're soon able to replace our bodies with robotic ones.
The remaining problem is to figure out how to keep a brain alive in a vat, and making the life-support devices small enough to fit inside the body. It's probably not too difficult; you need to circulate blood, remove carbon dioxide, add oxygen, add glucose. You could probably keep the brain alive indefinitely.
> This means that we're soon able to replace our bodies with robotic ones.
"We have a computer that can play chess, which means we'll soon be able to replace humans by AIs" - someone in 1970
> You could probably keep the brain alive indefinitely.
> It's probably not too difficult
Why do 1/3rd to half of people above 85 have Alzheimer's/Parkinson's/Dementia then ?
Software developers have very simplistic and optimistic views about these things, we're very far away from anything remotely close to have you're talking about. The human body is infinitely more complex than what your comment hints
My hopes are smaller in scope. I hope this helps people with severed nerve damage, or perhaps to implement better prosthetics for people with missing limbs (perhaps with motion and tactile sensors?)
One that I would personally be very interested in would be improving human-computer interactions. Preferably with no quirurgical procedures, or something extremely non-invasive and quick.
Helping disabled people is probably the most important near-term application of this. I just like to ponder about long-term "sci-fi" implications, in terms of how and when these kinds of ideas will become actually plausible.
Some applications, like digital eyes, which connect directly to the optic nerve, would be first used to cure blindness, but soon people would start replacing their functioning eyes too, if it allows feeding pixels directly into the brain. Same with prosthetics, if they'd be better than biological ones. Then if you have cyborg eyes, arms and legs, why have a biological 'core' which is pretty much useless at this point.
Storing the brain in the body would actually be a bad idea. Human brains could be stored in a warehouse and they'd control robotic bodies remotely. You'd be able to jump into different bodies all over the world, and choose from different form-factors (ant, whale, dragon, etc.) This would actually be easier to implement, because the life-support system for the brain could be larger and consume more energy and other resources. You'd also be able to choose whether you want to live in a simulated world or a physical world through robots.
I think this is where we're quite realistically heading in less than 50 years.
As these technologies become reality, funny to think that we may see a class of disabled people leapfrog the rest of society and become hyper-enabled. Becoming essentially superhuman, with heightened abilities in things like vision or prosthetics due to being the first to adopt bio augmentation
"This means that we're soon able to replace our bodies with robotic ones."
Come on man, I can't believe people do these wild extrapolations out of nothing like this, when every two days an article on HN and a hundred on reddit claim to definitely cure cancer and aids and whatnot
You can do this with atom level simulation and growing the brain all the way from DNA, but it needs a lot of compute (> 50 years).
However, and in the end you'd be bored, because you could get rid of all of your human flaws, which is what makes life interesting. It'd be really depressing and boring to be a superintelligent computer.
Never wish for this. The amount of (emulated, but still real to the emulatee) suffering has no practical limit. Emulation/Uploading is every sadists wet dream. The only way to stay free is to live your life in the real and embrace (and hope for) the permanence of death.
I've always found it incongruous that we can move a person's body around the globe at 900 kilometers per hour for 25 hours for under $1500 but for 1000x that price and even with 1000x that time we can't bridge the distance between a brain and a foot.
So, I'll put this into terms that a sysadmin can understand: imagine if all the Ethernet or fiber going to a datacenter rack was bundled up, and I cut that bundle up. Now you have to splice them back together so that every port is still connected to the same port it was before I cut the cables. Oh, and also, the cables are microscopic, incredibly fragile, and we don't know how to actually repair them. We sort of just hope they can grow back in place.
For scale, a human spine has coming off it a few hundred thousand nerve fibers inside 31 pairs of nerves.[1]
I was actually expecting more, since that is effectively IO for the whole body, and each individual nerve fiber tends not to fire more than tens of Hz.
it's all connected to an organic deep neural network
which means that the specific arrangement of the physical wires doesn't matter. because there's a training period in which the neural network literally learns to control the body anyways
There's no reason to be glib. If patients' brains can reconfigure following trauma such as strokes [1] or having our entire visual field flipped [2] there's no reason to assume one couldn't reroute around having nerves hooked up differently.
Can speak from personal experience. Traumatic incident which severed the nerves to my leg in multiple places. Nerves eventually regrew and reconnected within the leg, and then again where they were severed in the foot.
Motor and senory nerves reacted differently.
When motor nerves reconnected, I still couldn't contract the muscles and went through a series of steps to relearn how to use the limb. First I was trying to "move" the leg, but effectively the "IP address" for the leg was changed so my "move" signals were going to where my head thought the leg was instead of to the new connection. Instead, I would estim a specific muscle and "listen" in my head for where "noise" was coming from. That "noise" was the electric buzzing from the estim'd muscle contraction. Eventually, I learned how to concentrate to make a muscle contract, and many steps later (pun intended), I learned to walk again.
Sensory nerves didn't need a push signal, they're like a constant inbound feed when connected. When the sensory nerves reconnected, it is something you definitely notice. Going about your day, and suddenly you feel an jolt, like being shocked, and over the next few hours to days the area that is reconnecting is burning, stinging, feels like it is being crushed by pressure, and cold all at the same time. It was much more intense than when your arm falls asleep. The sensation can be maddening but it eventually passes as your body begins to sort and acclimate the signals.
All of these steps on calibrating the sensory nerves and learning how to contract and coordinate muscles is something we take for granted as people usually sort it out when they're infants.
I've had exactly these experiences following a spinal injury -- fracture of a vertebra but with minimal cord damage and quite a lot of disruption to the dorsal root ganglia.
You can't put into words how weird it is to fall over because your brain thinks your foot is somewhere it isn't. Or how suddenly you become incredibly aware of how the front of your calf feels. Or how overjoyed you are to be able to move your toes again for the first time in a year. It's not like what you see on the movies.
Wallerian degeneration -- yes, degeneration -- is part of the healing process of some grades of nerve injury. Things literally get worse before they get better, as the fragment left of the crushed axon degenerates to its root and then regrows. It's incredibly slow -- around 1mm/day at most -- and a matter of probabilities. What's also worth mentioning is that there are plenty of internal nerves too, where restoring function after a trauma would be life-changing -- like the Vegas nerve, which buggers up lots of things if damaged slightly, or, in my case, some of the nerves in the fundus and neck of the bladder, meaning that my toileting is really very different than it was before.
I'm glad you're doing better, and hope you continue to do so. I've no idea if the device the article is talking about will ever help, but nerve injuries cause so much disability worldwide I'm glad they're continuing to be worked on.
This and the parent comment really should be at the top of this page. They are the best descriptions of how this sort of thing works in practice I have encountered. And while they tell how difficult and slow it all is I feel that they could give hope to others that some sort of recovery can be possible even without new advances in treatment.
Amazed congratulations to you and parent. I can't imagine going through that, but then I imagine you wouldn't have chosen to either. Hope things continue improving.
Your description also made me reflect on infants, and whether we effectively "feel more" in that stage, as our nervous systems are self-calibrating and adjusting gain.
Did you use any medication to help nerves regrow, like Lion's Mane or something like that? We have plenty of injured soldier here, in Ukraine. I'm looking for something cheap and effective to help them recover from injuries.
Imagine that the bundle of cables were all completely identical and unlabelled in any way. Even if you could surgically reconnect the nerves perfectly, you don't know which nerves should be connected to what on the other side.
As another commenter posted, there are 31 "bundles" of nerves carrying a total of a few hundred thousand individual fibres. The odds of re-connecting a human spine correctly are simply incredibly small.
Is that interesting? It seems natural. Precision has always been quite hard. Most people can kick a ball 10 m more easily than they can kick a ball 1 m but to center the hit at a 1 mm spot. I can carry a pile of sand 100 m easier than I can move 1 grain of sand precisely 1 cm without disturbing the others around it.
Given enough curiosity, anyone can understand how and why a jet plane works in as much or as little detail as they want. Planes were engineered by humans from first principles and we're good at both documenting our inventions and understanding those first principles.
Building something to interface with a biological system, though, is another matter entirely. It could as well be alien technology. It requires reverse engineering a lot of extremely complex stuff that was not designed by our civilization. So incomprehensibly complex that we only fairly recently made enough progress in other fields to be able to build tools to meaningfully poke at it.
To the best of our knowledge, it wasn't designed at all. It's more like giving a monkey a typewriter and millions of years, with a very crude feedback function (evolution/natural selection) that makes it very hard to backtrack out of a local maximum.
It's quite hard to interface with it because it's not designed to be interfaced with and the whole thing works pretty much by accident, and where changing any variable could throw it off.
Obviously. There is no functional separation between components (some proteins do multiple unrelated jobs), there's no distinct hierarchy (layers just kinda flow into each other), and there's way too much global state that works seemingly something like this: https://www.youtube.com/watch?v=I5mwVv5NjhA
And biology in particular is unique because it requires other fields to progress pretty far for the tools necessary for comprehensive biological research to become possible. As in, a microscope is a prerequisite for the discovery of cells and microorganisms.
People think medicine is way more advanced that it actually is. We can't add back some crystals on a tooth. We can't re-attach a nail to a nail bed. We can't fix cartilage. We can't physically repair arteries (short of donor material) or varicose veins. List goes on and on.
The class of problems that falls under the category of movement of matter is fairly trivial.
Repairing severed nerves is more like an entropy-reversal class problem. I don’t think we could even put a sufficiently broken tea cup back together exactly as it was.
These seem completely unrelated. One is about travel distance the other is about forming tissue connections. We can easily travel the distance between brain and a foot, it's around 5 ft.
Well if you think about it. The microscopic scale inverted in the other direction is more like getting to mars and the complexity is the same as creating a self sustaining colony there.
It might help to remember that a human body has more cells than there are people on Earth by around three orders of magnitude, and that you're engineering things on the scale of nano- or micrometers.
A typical human cell is on the order of 10 micrometers. If you need to bridge even 1 cm of that, you're bridging ~a thousand cell-widths. If you think of a cell as the somatic equivalent of a house in a city, that's the equivalent of an infrastructure project spanning (based on a quick count of the number of houses on each block in Oakland) the equivalent of around six miles, or roughly from downtown Oakland to El Cerrito on a map of the Bay (~4 BART stations). And you have to do that on a scale where precision manufacturing is incredibly hard, where you're dealing with extremely difficult problems of chemical synthesis, in a living body, without provoking the body's defense or repair mechanisms to stop you. And that's assuming you even know what you're trying to do, which requires an understanding of the machinery of those cells that we often don't have.
I wonder if we could augment/cross our nerves to control things we normally can't control? what if we could release hormones on demand, like maybe release adrenaline or calm down?
Reminds me of a few hilarious plot points in a few Ian Banks novels. One of them, the Hydrogen Sonata had this passage:
“Is it true your body was covered in over a hundred penises?”
“No. I think the most I ever had was about sixty, but that was slightly too many. I settled on fifty-three as the maximum. Even then it was very difficult maintaining an erection in all of them at the same time, even with four hearts.”
Silly as this is, Ian Banks had a way of taking the mere hint of the possibility of a thing to it's logical extremes. His characters change sex, regrow limps, or morph themselves into a different alien species basically in an orgy of hedonism and utopianism. Definitely science fiction/fantasy when he wrote it but eerily close to becoming science fact as time progresses.
I think the spine is far more complex than the average person is aware of. From what I have read it has some basic reaction/thinking capabilities which may be beyond what can be made by humans.
University hosted announcements (e.g. rice.edu announces a major breakthrough) have incentives that are so out of wack it's basically clickbait at this point.
Researchers oversell their results for publishing and funding purposes. Then the university oversells those results to draw in students/investors.
Curious if others feel the same way or if I'm just too cynical at this point...
I understand the cynicism. Academic reporting is a hard & thankless gig.
The non-cynical version: Half of a researcher's job is disseminating results -- to other researchers & the populace at large. I'd much rather have excited university announcements (even over-reaching ones) than the rage-porn news that dominates the airwaves & "mainstream news" homepages.
As someone who has peripheral knowledge in this space (piezeoelectric & magnetostrictive devices and neural stimulation & recording): I've never seen a magnetostrictive material capable of developing a DC bias. Further, driving the device in a non-resonant mode for neural stimulation is even more new to me -- that's quite fascinating, and I'd say that article lives up to the hype from my 1000-ft vantage. I.e. This was effective reporting.
30 % of physics and chemistry can’t be replicated. 70 % of soft science can’t be replicated. Most of what we know about sociology may just be fake, and given that a huge foundation of progressive thought around rase, sex, and gender is based on this, this is a big problem.
> 30 % of physics and chemistry can’t be replicated.
Is this true? I left academia a long time ago, but I'd be surprised if it's that good. I'd actually suspect it's the inverse and only 30% can be replicated.
Considering how hard it is to actually fix any accidental damage to brain cells and nerves, and the far-reaching consequences of such damage, I am always amazed to see how many are eager to just plug something into it and let some arbitrary code send electric shocks through their brain.
I get the unfathomable potential of the thing, and the appeal of said potential. But that’s a product I won’t be an early-adopter of.
Yeah, every time something like that is mentioned, I just think of small issues that could cause noise on the output. Basically if you wear some device that helps you, either passive like glasses or active like a hearing aid, you can immediately disconnect then if needed. But imagine an implant bugs out and you start receiving a maximum signal for something without being able to stop. Like a blinding bright light which doesn't go away when you close your eyes. As much as the cyberpunk idea is fun, I'd have to really suffer without some implant to risk it.
I wonder if it's possible to make a much simpler passive-only "mental stress detection" system, and use that as a kind of emergency stop (let's not call it a "kill switch" in this context) for any more-complicated augmentation.
That creates the opposite risk that your enhancements stop working when you _want_ them in some kind of real/external emergency situation.
I suppose it also wouldn't work in the case that the main augmentation caused, say, unending waves of paralyzing calm, although there could be another detection-gate for physiological signals to keep such people from starving/dehydration.
Not the parent, but I vaguely recall our ethics teacher mentioning some associations advocating for hearing impaired people have, at least in some European countries, advocated against cochlear implants in young children, mostly on ethical grounds.
His whole point being that ethics is not about what some majority might instinctively believe is best for everyone, and that other perspectives must be considered.
