Growing a Brain in Switzerland
February 21st, 2007If you liked the one about the bits of rat brain and the flight simulator, you’ll love this!
Via: Spiegel:
A network of artificial nerves is growing in a Swiss supercomputer — meant to simulate a natural brain, cell-for-cell. The researchers at work on “Blue Brain” promise new insights into the sources of human consciousness.
The machine is beautiful as it wakes up — nerve cells flicker on the screen in soft pastel tones, electrical charges flash through a maze of synapses. The brain, just after being switched on, seems a little sleepy, but gentle bursts of current bring it fully to life.
This unprecedented piece of hardware consists of about 10,000 computer chips that act like real nerve cells. To simulate a natural brain, part of the cerebral cortex of young rats was painstakingly replicated in the computer, cell by cell, together with the branched tree-like structure of the synapses.
…
The newborn “Blue Brain” surprised the designers with its willfulness from the very first day. It had hardly been fed electrical impulses before strange patterns began to appear on the screen with the lightning-like flashes produced by cells that scientists recognize from actual thought processes. Groups of neurons started becoming attuned to one another until they were firing in rhythm. “It happened entirely on its own,” says Markram. “Spontaneously.”
…
Skeptics wonder what the purpose is of painstakingly replicating things when scientists have so little understanding of their purpose and function. Indeed, no one will know, within the foreseeable future, what exactly happens in the circuits of the replicated brain — except that whatever it is, it looks seductively authentic to an outside observer.
Perhaps the only ones who could have known are the scores of rats that carry on a shadow-like existence in this supercomputer. Researchers opened thousands of rat skulls over the years, removed their brains, and cut them into thin slices, which they kept alive. Then they directed tiny sensors at the individual neurons. They listened to the cells firing neurons, and intercepted the responses coming from the adjacent cells.
The brain slices were exposed to a variety of electrical impulses. The impulses reflected the stimuli that may have been received by the laboratory rats’ brains when the animals smelled cheese or were startled by a shape. The cells reacted — just as they did when the rats were still alive — by sending electrical charges through the neuroplexus. The researchers’ measuring devices recorded all signals until the brain slices expired.
In the end, the researchers at Markram’s lab collected the entire repertoire of behavior of hundreds of types of cells in every conceivable situation in a rat’s life — stored in endless tables. This vast stockpile of data let the researchers start building their digital doppelgänger.
Research Credit: Life After the Oil Crash
throw in synthetic free thinking brains, add a touch of the law of unintended consequences, sit back, and enjoy.
actually, the intended consequences will be bad enough.
s
Could this be the birth of something akin to Skynet or the master robot in I-Robot, where rudimentary A.I. slowly manifests itself throughout our civilization, developing and learning about humankind so that when it truly does become self aware, realizing humans are a menace to the planet and itself, and it is so well integrated in our society and technology/military that we have no real way of controlling it or stopping it, or am I just full of scifi-esque maniacal delusions…
BG,
I wish my buddy would step in here. He has a PhD in cognitive science. He was working on a project that he acknowledged would someday wind up in the brain of a terminator robot. His adviser used to tell students, “Let’s face it. We’re building terminator brains here.”
At his dissertation defense, he told the panel something like: This research is evil. We shouldn’t be doing this. Someday it’s going to come back and bite us all on the ass.
He was implementing, in software, the universal neural structure that exists in the brains of ALL creatures that have brains. Cockroach brains. Dog brains. Dolphin brains. Human brains. At this level, it’s all the same. What mainly varies are the concentrations of neurons gathered around the various sensory input centers, and the overall number of neurons.
Get this: He’s convinced that what they’re looking for isn’t in there in any neurophysiological sense. He said that there’s no path to understanding human consciousness via modeling the thing in that way: But that doesn’t mean they won’t be able to make it kill.
It doesn’t have to write poetry, or have feelings. Specific neural processes will be modeled, based on data gathered via the wet work on the rats and other creatures. Like the article said, they record all the data so they can use it in their supercomputers. The data collected off the clusters of live neurons themselves… That is what will lead to the shocking, sci-fi stuff.
Even though he only developed the “system” to function in a theoretical capacity, my buddy said the thing did weird, unexpected things. It amazed and frightened him. His project was only using a few gigs of ram and 4 CPUs. Gaze upon that thing in Switzerland…
What did my friend with the PhD wind up doing?
He lives in an RV and wants nothing to do with academia. He’s doing programming and enterprise linux stuff now. No terminator brains anymore. He’s saving money to buy a couple of acres of land to park his RV on, and to keep a small garden and some chickens.
He sounds like a smart man.
That totally makes sense. As far as the military goes, they ultimately would have one goal in mind, and that would have to be the super soldier, the ultimate killing machine that neither feels emotion but views killing as just another task. Looking at current age video game technologies A.I. it would seem to be a logical next step to marry this type of knowledge wilth a mobile gun carrier/robot killer to do all the dirty work.A machine with a uncanny ability to know and respond to a threat before a person even has a chance–this all will happen once robotics hardware catches up with software, we’ve already seen it with the monkey controlling the flight simulator with his mind, or the man with no arms controlling robotic arms with sensors hooked up to his brain. They can mask this research under the guise of helpful medical research to assist disabled people, but I fear a more sinister purpose may be planned.
I’d have to agree with your friend with the PHD and switch to doing research on something more beneficial to humanity..or just get off the grid,. I know i definitely wouldn’t want to be responsible for this type of “smart” weaponry.
>>>I’d have to agree with your friend with the PHD and switch to doing research on something more beneficial to humanity.
That was his goal, going in. But it’s the old, follow-the-money routine. Everywhere he looked, the thing was just awash in military money. I started seriously looking at the number of universities that were doing this work, and the money behind it. What I found was so shocking that I wanted to make a full length documentary film about it!
Life kinda got in the way for me, but I hope someone makes that film someday.
Kevin
I am working on my PhD in cognitive neuroscience, and the lab that does modeling work here is actively seeking funding from certain DARPA projects whose explicit purpose is to create autonomous and artificially-intelligent war fighters.
It can be OK to work the system from the inside, but it is certainly frightening.
You guys should check out Moreno’s recent book “Mind Wars.”