Slashdot Mirror
Nerve Cells Successfully Grown on Silicon
crabpeople writes "Researchers at the University of Calgary have found that nerve cells grown on a microchip can learn and memorize information which can be communicated to the brain. 'We discovered that when we used the chip to stimulate the neurons, their synaptic strength was enhanced,' said Naweed Syed, a neurobiologist at the University of Calgary's faculty of medicine."
But what's the size of a neuron vs the size of a transistor in a 65nm process CPU?
-- *My* journal is more interesting than *yours*...
"We discovered that when we used the chip to stimulate the neurons, their synaptic strength was enhanced, ... "
There was something like this in one of Asimov's books. The guys synapses are enhanced by a machine, then the guy starts to "feel" and "manipulate" things.
But this is very exciting. The idea that we could grow neurons on silicon is one of those big steps that looks to lead us into the Johnny Mnemonic world that Gibson was talking about just a couple stories prior to this one.
There is a song that says, "It only takes a spark to get a fire going". So too is it true that it only takes a couple neurons to start synapsing. As these true neural webs become more complicated, it would be interesting to see if any kind of emergent behavior was evident.
Also, with the current political and scientific climate as it is, this could be the first step to replicating a nervous system without having to rely on fetuses for stem cells. It requires no human cloning and holds immense promise.
It would definitely be cool to have a couple of these chips implanted to enhance the base memory that we are kitted with at birth, that's for sure!
I have been pwned because my
I thought the Pine Lab at Caltech had done this several years ago. Neurochip Project
Will this make computers more human or otherwise ?.
... Forty Two ... naah... doesn't work) ... Of course, nature did a better job making us humans than we would have achieved ... :)
Maybe it's time to admit that nature does a better job bruteforcing (OK , what else do you call SEX and EVOLUTION) the secrets of this world than all our mathematical precision.. (E=MC2
Quidquid latine dictum sit, altum videtur
If only they could find out how did the strength increase and wether we can do the same to the human body we can find a cure for most of the nervous system degradation diseases. Anybody have link to a more verbose article?
My Aurora : http://www.youtube.com/watch?v=o91ZsGwJYyg
FB : https://www.facebook.com/TanveersPhotography
Not making faster Pentiums or Athlons. Sorry. Most of that magic has already been woven. Who out there is qualified to make systems level designs and decisions about bio computer systems? Think about the type of knowledge it must take about physics, electrical and computer engineering, as well as biological knowledge.
What type of magnetic and power restrictions will there be? Reliability? What type of optimizations will exist? Interfaces? Flexibility?
We're still quite far away from having things like this be applicable to modern day but think about when you too can say, "I know Kung Fu"!
The researches have read some Slashdot posts, and believe that there must be a huge market for this chip. There is clearly a need for it ;-)
.. memory upgrade implant, specially in the mornings.
:-)
It would also be cool with an encyclopedia or even a few o'reilly books implanted.
Too bad it seems to be a one-way communication only, otherwise a spellchecker implant would be cool too
Researchers at the University of Calgary have found that nerve cells grown on a microchip can learn and memorize information which can be communicated to the brain.
While the article mentions this in the introduction, it doesn't mention this happening at all in the research. It talks about neurons communicating with each other. This is a long way from connecting this chip into a living brain in an animal that can still function.
While I agree that this is a fascinating article, we should make sure not to sensationalize it too much. Making chips that interface with actual brains in actual animals, even if they are snails, is still a long way off.
"Flying is the art of throwing yourself at the ground and missing." - Douglas Adams
would have a whole new meaning...
...but I still think Natural Stupidity will outpace Artificial (or artificially enhanced) Intelligence.
Ely Lilly release the new Prozac add-on for nervous cpu's
a.k.a Neal Stephenson and his uncle.
Chip embedded in politician's brain after a stroke - he goes on to be president.... v. spooky.
I would love to see alzheimer's patients helped with this. If it's a genetic disease, I'm up the creek and dropped me paddle a while back.
- Lnr
Evolution != bruteforcing. With bruteforcing (e.g. trying to guess a password with a dictionary) there is no "being on the right path" or whatever. It's just wrong or right. Evolution is survive of the fittest, do minor changes in different direction on an existing system and let see which one will lead closer to success.(Just like sex ;-)) Take many of the fittest and do the same again. The some time take some of the not so fit and try as well the same.
On the other hand you are right: This trial and error seems to lead to better results in the long run compared to deterministic creation. But this scheme is already adopted by science. IIRC there was a distributed computing project simulating a robot with a defined task and changing the parameters of the robot. The different clients exchanged the information about the results. I don't remember anymore the name or the homepage of the project, I think it was already 4 or 5 years ago...
Trolling is a art!
Imagine a U.S. President that is simply a marionette made of organic plasma being controlled and manipulated by puppeteers and handlers behind the curtain - stringlessly AND wirelessly.
towards a virtual girlfriend.
Quoth the article:
scientists stimulated one nerve cell to communicate with a second cell which transmitted that signal to multiple cells within the network.
Singal up (probably down too, though that is not said). That's a start. Now let me jump.
Imagine how this would feel in your own brain. Even strengthened to noticeable level by a lump of neurons, the signal would still read "beep". Now imagine being fed information through that channel. "Beep, bip beep bip bip beep". Better start training that morse.
Now let's enhance the input by adding more bits into it and running data through a digital-to-analog converter. This is where you would slowly be able to "see colors", one at a time. Low signal, cold feeling; high signal, hot feeling. That is brainable information. You can associate different patterns of these "colors" to different ideas.
But still it's not like you could see any shapes, is it?
Now add more bytes, feed them in side-by-side. That's a feed. At this point, feel nausea. Something is feeding noise into your thoughts, something you cannot possibly comprehend.
Would take a processing system not unlike vision inside the brain to translate that feed into experiences like colors, tastes, touches, then further associate these to make shapes out of the noise.
A long way.
Worth taking, of course, as research goes, but I wouldn't toss away those external displays as of yet. Have a hunch computers won't be the same, either, when we get there.
Future research will focus on interfacing silicon chips with the human brain to control artificial limbs and develop "thinking" computers.
Mostly fun!
I think, therefore thoughts exist. Ego is just an impression.
Alan Cooper, author of "The Inmates are Running the Asylum" and other texts put it this way:
Q: What do you get when you cross a camera and a computer?
A: A computer.
His point is that from an interface and place-in-the-world point of view, most products that have been digitally enhanced tend to remain closer to their technology roots than their analog counterparts (with all of the usability, and I would say ethical, challenges inherient in a technologist-driven system).
That said, this is pretty frickin' cool, but the double-edged sword presented by this innovation seems both particularly sharp and far reaching. I really hope we get this one right.
"Why can't you use your powers for Good?"
So that's how the "real people personalities" work. Guess the crowd at University of Calgary will be the first against the wall when the revolution comes!
An eye for an eye... leaves the whole world blind.
A quote i read somewhere
"The danger from computers is not that they will eventually get as smart as men, but we will meanwhile agree to meet them halfway." -Bernard Avishai
Tinnitus is a serious problem to a lot of people today, and it can have many causes, from various diseases/illnesses, to noise damage. It apparently has to do with the nerves in one's ear, so would this kind of research, might we finally see a way to actually treat tinnitus?
Until you get T, you don't realize how lucky people who can actually be in a quiet room without going mad are...
Clever signature text goes here.
Pubmed link to the abstract for their research. Publisher's site sometimes holds a free copy of the full paper (depends on the journal).
Neurons are much larger than transistors, but the two aren't really comparable. The main body of a neuron is usually around 25 microns (25000 nm) in diameter and runs at a clockspeed only in the kilohertz max.
A neuron is much more than a transistor-like switch. On the one side of the neuron's central body is a set of dendrites that connect to and gather input from other neurons. The average neuron might have a thousand of these dendrites.
The synapse at the end of each dendrite acts like part of a multiply-accumulate term -- taking the signal from an other neuron, multiplying it by a numerical coefficient and summing it into the total excitation level of the neuron's body. I suspect that the precision of this multiply -accumulate process is fairly low -- perhaps 8 to 16 bits.
Next, the body of the neuron has a long axon extending from it that sends the output of the neuron to other neurons (connecting to the dendrites of other neurons). This axon can be quite long, millimeters, even inches, in length. Thus, the axon is like an off-chip line driver with the potential to have a very high fanout (of a 1000 or more). (On a modern microchip, these off-chip connections are driven by much larger transistors than the small 65 nm ones used in computation).
Third, a neuron is not a static multiply-accumulate system. The coefficients on each synapse change in response to long-term adaptive processes. This process is computationally complex and includes cross-correlation of inputs between synapses and processing of other chemical signals in the brain. Cross-correlation alone could require the equivalent of several kilobytes to several megabyts of RAM. (We won't even get into the adaptive processes that include physical growth and removal of dendrites as this has no easy analog in hardware)
In summary, a neuron is more than a transistor-like switch. Its a free-running 1000 register multiply-accumulator with an off-chip line driver and a statistical processing engine that updates the coefficients on each of the multiply-accumulate terms. Thus, emulating a single neuron would require hundreds of thouands to millions of transistors.
Two wrongs don't make a right, but three lefts do.
Thanks for the reply, very enlightening.
But it clearly would be folly to try to emulate a neuron using purely digital computing techniques. You're dealing with an analog mechanism that is pretty much a wire-or of many inputs feeding into a capacitor. This is very much an analog computing circuit; now the question is how efficiently you can do A/D-D/A conversion on this scale.
(And as I recall, the sciatic nerve running down your leg is a single cell with an axon over 1 foot long. Definitely some impressive stuff Mother Nature has concocted...)
-- *My* journal is more interesting than *yours*...
Or, for a more software interpretation, it's a function that takes a bunch of boolean parameters and returns a boolean. Anyone who's ever done any programmation or computer architecture should see why you can easily process anything with this.
This axon can be quite long, millimeters, even inches, in length.
Acutally, it can be over a meter in length (spinal cord to calf is one axone). Try that with a transistor
Ask 8 slackers a question, get 10 awnsers (a citation, but I can't remember from who)
Since you are using a ridiculously little part of all the storage space you have available at anytime in your life. The cool thing, would be to have an electronic device that could strenghten a give synaptic path, allowing you to "refresh" your memory at will, and not forget important things. (like read the whole C++ W/ libraries reference once and then refresh this everynight)
Who'll be the first to upload a linux distro into the brain of an actual pinguin.
Or, for a more software interpretation, it's a function that takes a bunch of boolean parameters and returns a boolean. Anyone who's ever done any programmation or computer architecture should see why you can easily process anything with this.
Excellent point. You are right about the computational flexibility of neurons. They can represent a wide range of logical functions, although I believe that the single neuron is incapable of doing an XOR.
But a neuron is more that a Boolean circuit. Although a neuron seems like a two-state device (its either quiesent or its firing), it is more of an N-state analog device in which the pulse-rate encodes a numerical quantity (probably the equivalent of an 8 to 16 bit floating point number). That is why the dendrite field is like a giant numerical multiply-accumulate.
Two wrongs don't make a right, but three lefts do.
Who cares? Just so long as I can use this to learn kung-fu (or how to fly a helicopter) in less than 10 seconds.
IAAN, and this is not a big breakthrough in any sense. Basically, this is something that was first done using manually-positioned electrodes probably twenty years ago, and now they can grow neurons on a dish that has electrodes built into it and do it that way. WoO-hAH!
The computational power of neurons comes from the way they work in groups, not the way they work alone. Therefore, it's strongly dependent upon the detailed organization of their connectivity. Grinding up a piece of brain and regrowing it on a dish will obviously not retain native connectivity. Additionally, the time it would take to manually rewire an interesting circuit by giving little localized electrical pulses (or do anything else interesting) is longer than neurons are viable in culture, and that's not a problem that's been solved yet.
I'm not saying this technology won't have important uses as a research tool, just that it won't be useful for what people here seem to think it will be useful for (high-density pornography storage). BTW, one of the more interesting characters in this field is Steve Potter, a somewhat strange guy who does some technically impressive work
You're right on-- the change in firing rate relative to the baseline firing rate is very important. Also, there is some reason to think (logically and biologically) that some ensembles of neurons fire synchronously with each other and asynchronously from other ensembles of neurons. By using synchrony of firing, they gain computational power and allow for variable binding, thus allowing more formally logical computations to happen than just autocorrelation our boolean operations.
If I have four neurons, and one represents "red," one represents "green," one represents "square" and one represents "circle," then it is very difficult to tell (based on the sustained activity of the neurons alone) whether they are responding to a red circle and a green square or a red square and a green circle. This is called "the binding problem" and, at least in neural networks, can be solved by distributing the firing patterns of the neurons over time. So, "red" and "circle" fire in synch, then rest while "green" and "square" fire in synch and then rest while "red" and "circle"... etc. Notice that you could even have "red" bound with both "circle" and "square" by being active over two epochs, thus allowing for dynamic binding of variables, etc.
Anyway, the point of all of this is that if we can figure out how some of this temporal synchrony dimension is exploited in the brain, then we should be able to harness that computational power through silicon transistors like the one described in this article and build modules that could replace damaged regions of the brain.
The term "outside the box" is squarely within the box at this point.
I think parent (along with some other posts) are confusing the biological neuron and the perceptron, which is a simplified mathematical model. While the perceptron can't cope with linearly inseperable problems (like XOR), there is no consensus on the computational limits of the neuron. In fact, very little is known for certain about the learning algorithm used by the nervous system. The neuron may learn not only through the weights of its inputs, but also through chemical interactions with glial cells. Really, the neuron is still too much of a mystery for us to know its limitations.