"Talk to me Goose, talk to me!"
Alright, not really...but almost, sort of.
Thomas DeMarse, 37, an assistant professor of biomedical engineering at the University of Florida, has been working on training Petri dish-grown rat brains to fly an F-22 fighter jet simulator.
DeMarse and his fellow university scientists first grew a rat brain by extracting neurons from a rat embryo. The neurons then mulitply, and voila, soon you have 25,000 neural cells ready to take on Iceman and win the Top Gun competition, right? Not quite. They keep the rat neurons in a Petri dish with special liquid to keep the neurons alive. Soon, the neurons clump together, form neural networks, and become a real working mini-brain.
The 25,000 cells are on top of a grid of 60 electrodes inside the dish. "These electrodes allow us to literally listen to the 'conversations' among the neurons to find out how they are computing," DeMarse said (pictured right with one of his rat brain dishes). "By sending in [electronic] pulses to each electrode, we can also stimulate the network in 60 different locations."
DeMarse then connected the rat brain to a jet flight simulator via the electrode grid and a regular 'ol desktop computer. They activated the brain, turned on the flight simulator, and let the rat noggin do the rest. I feel the need for speed! And some cheese!
How well did the brain do?
"When we first hooked them up, the plane 'crashed' all the time," Dr DeMarse said. "But over time, the neural network slowly adapts as the brain learns to control the pitch and roll of the aircraft. After a while, it produces a nice straight and level trajectory." The brain even learned how to keep the jet fighter straight in mock hurricane force winds.
Does this freak anybody else out???
So what's the purpose of such experiements? Are we going to crush Iran with a squad of rat brains or something?
"We're hoping to find out exactly how the neurons do what they do and extract those rules and apply them in software or hardware for novel types of computing," DeMarse said. In other words, this research could possibly lead to the creation of sophisticated, real thinking computers. Imagine - a computer that could actually think, be creative, and be flexible enough to figure out more complex and open-ended problems. Even the most powerful computer lacking the ability to "think outside the box" wouldn't know the differnce between a dog or a cat if it had no previous knowledge of either. Giving a computer a "biological" component would enable it to figure it out. This type of thinking is what we humans take for granted, but it's currently impossible for a computer to do.
Of course the research has military implications. One day, they could install living computers in unmanned aircraft so they can be deployed on missions too dangerous for humans. This work also has medical breakthrough potential as well. Studying the nature of neurons might provide the basis for developing new drugs to treat brain diseases such as epilepsy.
The U.S. National Science Foundation has awarded DeMarse and his group a $500,000 grant to produce a mathematical model of how the neurons compute, and the U.S. National Institute of Health is financing research into epilepsy.
I see a lot of potential benefits from this research, but I can't help but wonder about the inherent risks involved. What if we take the next step and see if lab-grown brains can fly a real jet? I think that would be very dangerous. It could put thousands of pilots out of a job. What about the creation of thinking and self-aware computers? They could enslave humankind and take over the world! OK, sorry...I must be paranoid.
How dangerous could self-aware, living computers be? Seems harmless to me.
Or maybe not...