Slashdot Mirror
Working Nerve Chip
poetic writes: "Two scientists from Munich have succeeded in creating a nerve chip with silicon and snail nerves. The cells were hindered from growing away from the silicon with a plastic fence. They managed to get a signal to go from silicon through a neural circuit and back to the chip again. Cute, one step closer to a decent uplink! See the
abstract at Nature's site."
What I'd really like would be to have a CPU in my arm. Not for real thought -- too complicated just now -- but for the odd maths test it'd be extremely useful...
Of course, it'd have to be something slow -- could you imagine a human with cooling fans stuck all over them?
You will be assimilated...
resistance is futile!
You're using her as bait, Master!
It is certainly a nice start to what could be a promising field, but there are so many things to be dealt with before we could realistically use any technology like this for clinical purposes (ie: borg implants). Namely, if these were to be used in longer nerves (anything in the spine for example) they would have to come up with a way of dealing with myelinated nerve cells, not just bare cells. Myelin is a sheath that covers nerves to increase speed of signal transduction, and piercing it (with a chip interface) could lead to problems in propagating an action potential... However, this stuff looks like it has serious promise for starting research into "biological computing"
... not in here, pal, this is a mercedes...
I saw a documentary on TLC a while ago about AI and such, and they mentioned a group of scientists who had done something similar to this. They were pretty vague about the whole thing but these people had basically taken a small clump of nerve cells (I want to say they were human brain cells, but I'm not sure...) and put them on this chip that would monitor their outputs and provide inputs. They had connected the whole thing to this computer which simulated a very simple 2D (pseudo-3d, kinda like Wolfenstein) environment, and trained the cells to move around in the virtual "world", avoiding walls and obstacles.
It's sounded pretty far out...has anyone else heard of anything like this?
--Ariston
"I'm never wrong--sometimes reality just disagrees with me."
I remember reading about this experement, and what they had done is taken the "brain mass" from a lamprey (a jawless fish with an incredibly simple nervous system). Pretty interesting that just a little bitty hunk of cells could navigate around in the computer. Since a lamprey has but one very simple "eye" it was probably fairly simple to feed in the visual input.
Regards
James
The idea of connecting my brain to the net is fascinating, but I can't imagine it being a usable technology within my lifetime. We still don't really have much skill with artifical neural networks, let alone the knowledge it would take to figure out how to get abstract information into a human brain. . .
In the near future, I see this technology as being more usable in fields related to bionics; I don't know a whole lot about the capabilites we could get from a microchip being linked directly to neurons, but I can imagine that it would provide a person much greater control of artifical limbs, and help a lot with artificial sensory organs. We have a rudimentary understanding of how the visual cortex works, so regrettably I have a feeling that the first widescale applications of this technology will be in attempts to link digital cameras to the brains of blind people or the like.
Even when we finally figure out how to get abstract ideas into peoples' heads, I imagine something much greater than the Net - I am thinking of a system where all the knowledge of the world is available to a person in a similar way to that of their long-term memory, so that if I wanted to "remember" how to use some obscure API call or somesuch, a hundred years in the future, all I would have to do is think about it, and the chip would link into a network of somesuch and pump the knowledge straight into my brain, as if I had always known what it is I was trying to find out.
I'm actually surprised at the fact that the article focused on prosthetic implants (ie into the brain) as one of the primary benefits of this technology. Many of us on /. followed with similar thoughts. However, I see other applications to be much closer on the horizon. While we have a long way to go before we can start wet-wiring silicon to our brains, I think we could use this technology for artificial limbs much sooner. Theoretically, it should be possible to build entirely eltromechanical limbs that have the ability to transmit feeling -- hot and cold, pressure, pain, etc. Being able to connect electronic sensors in these limbs to actual nerve tissue is the missing link. In cases where a limb truly could not be attached, this would be the next best option. Any speculation on how long it will be?
GreyPoopon
--
Why is it I can write insightful comments but can't come up with a clever signature?
If this has your interest, UK professor Kevin Warwick definatly will. He had a chip implanted in 1998, making him a cyborg (not the first though). He researches robotics and decided to get the implant for a week or so, which communicated with the university where he worked through a radio link (his story in Wired).
;) is that he is doing this on himself. There are actually health risks involved in the operation, which is why he chose to get it in his left arm, as he is right-handed. I guess someone would argue that it's unscientific experimenting on yourself, rather than a test subject, but for me it shows how much he burns for this subject, and if see an interview with him, I think you will agree.
THIS year, he is taking it to the next step. "Project Cyborg 2.0":
This phase will look at how a new implant could send signals back and forth between Warwick's nervous system and a computer. If this test succeeds with no complications, a similar chip will be implanted in his wife, Irena. This will allow the investigation of how movement, thought or emotion signals could be transmitted from one person to the other, possibly via the Internet.
I heard on BBC, where he was interviewed, that he wanted to find out, if they could transfer/share pain, he and his wife. Interesting stuff.
What I personally find cool about Kevin (yes, he is a first name kinda guy
-Kraft
Live and let live
His next experiment was similar but involved attaching the transponder to the epineurium (sheath) of one of the nerves of the arm- the idea was that the transponder would pick up signals (eg the axonal activity caused by touch sensation, or pain) and then that these signals could be sent to a computer and encoded as "patterns" (eg one pattern for holding a pen, one pattern for being pricked by a pin). These patterns could then be analysed and even sent back to the transponder, where it could now act as an output device, and cause the sensation that was encoded! There was even talk of implanting Prof. Warwick's wife with an identical transplant and putting them in continual communcation, so that for example, when Warwick stroked a kitten, his wife would hae the sensation of kitten-stroking.
Not surprisingly for those of us that have neurological/neuroscientific training, the results from this study have never seen the light of day. The ideas are flawed from top to bottom. Warwick's main mistake is that his second experiment has no relation to the first. The first (having a transponder that identifies individuals) is marginally interesting, if overblown (the transponder doesn't have to be surgically implanted to work) - his idea is "The chip implant technology has the capability to impact our lives in ways that have been previously thought possible in only sci-fi movies. The implant could carry all sorts of information about a person, from Access and Visa details to your National Insurance number, blood type, medical records etc., with the data being updated where necessary."
Im sure fellow
Anyway, that "experiment" (more like a beta test) doesnt logically lead to the second nerve implant. His lack of elementary neuroscience is evident here- peripheral nerve trunks are not good places to encode data- and if he did manage to "record" patterns for himself - how could he "play" them back on his nerves? A simple magnetic transponder? It would be like trying to email a GIF to someone by holding an industrial elctromagnet next to a bundle of phone-wires! And the thought that recorded patterns could be played back on another person's CNS using such crude technology is simply unbelievable.
Professor Warwick is regarded as something of a quack in the UK high-tech/neuro community, as this site, Kevin Warwick Watch, testifies. His research, however, does raise one or two interesting questions. His techniques and methods, though, are nopthing more than circus sideshows, compared to the excellent work with the mollusc neurons.
Tonight the sky is empty. But that is nothing new