Slashdot Mirror
Hybrid Robot Uses Rat Brain
CowboyRobot writes "After two recent stories of artificial brains used to control rats and one about MIT doing the reverse, NYTimes now has a piece on similar work done at Georgia Tech From the article:
"...the layer of rat neurons is grown over an array of electrodes that pick up the neurons' electrical activity. A computer analyzes the activity of the several thousand brain cells in real time to detect spikes produced by neurons firing near an electrode." But this time you can buy one for $3,000."
Whatever happened to the stereotypical guniea pigs? I think we should put their brains in robots, and see what happens.
When you don't have a leg to stand on, don't even get up.
If you could create a multi-laminar structure, this setup might be ideal for an artificial retina. Currently, the bionic retinas being used are nowhere near as sensitive as they need to be to create any useful phototransduction, even if the neural retinal substrate underneath remained intact (which it does not). A multilaminar device could sandwich photosensitive elements combined with neural substrates that would function as the neural interface to the output of the retina, the remaining ganglion cells.
Visit Jonesblog and say hello.
This rat-to-robot or robot-to-rat research thing is strange. Two things spring to mind :
- Isn't this rat brain interfacing business just a clever way of saying "ahem, moving right along" after decades of general-purpose AI research failure ?
- What the hell do these people target rats that much ? don't mice do the trick too ? or cats or dogs ? Some years ago, bio-computer interfacing experiments were conducted with squids, because they have very large neurons that are easy to work with : have squids complained to the PETA ? or maybe some of these researchers have pest have family members who work in the rats control business.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
How do they keep the nerve cells alive? Are they actually fed, oxygenated, and protected from infection?
yep. thats the problem with all this research....everyone who does it doesnt share their results. wheres the models for the function reponse of the rat neurons ? the electrical interface to the cells ? the procesedure and problems encountered ?
By the time anyone publishes results its years and progress has already moved on. the scientific system should be overhauled methinks. this research is critical and interesting enough that lots of people would be ahppy to contribute significantly if it was easy to obtain. a coupla thousand geeks playing with biological-electronic hybrids could do more than a bunch of researchers at a single university or two.
In other news, Soylent Green is made of people!
This guy came to talk at my college a few years ago, and his research is super-cool. At the time he mostly talked about trying to interpret what the cell firing patterns mean when the cells are totally isolated and then began adding single inputs, etc. Its neat cause when you cut them off from input they go into this pattern of waves of firing - they'll be these pauses with one or two random firings and then a all the sudden they'll all fire for a few seconds before dying down. He proposed that this was just like how a normal brain freaks out during sensory deprivation. Good to see that he's expanding on with some really badass research. This has the possibility of learning a lot about how neural connections grow in response to input. Plus it sounds kinda neat to have a rat-neuron driven robot. Now if only we could hear from that doctor studying human balancing to make better robots.
...was done by artist/writer Mike Saenz in his 80's comic "Shatter". He even had a diagram of "bug-bomb" robots with rat brains for processors because they were cheaper to use than microchips.
Also of note: "Shatter" was the first commercial comic book to be produced by computer (the Mac to be specific).