|Advertising|Jobs 転職|Shukan ST|JT Weekly|Book Club|JT Women|Study in Japan|Times Coupon|Subscribe 新聞購読申込|
|Home > Life in Japan > Environment|
Thursday, May 2, 2002
Robo-rats galvanized into action
Free animals guided by remote-control stimulation of brain
By ROWAN HOOPER
When the Italian physician Luigi Galvani published his theory of "animal electricity" in the 1790s, it roused biologists and physicists all over Europe, went on to influence the construction of the first electric battery and inspired an 18-year-old English girl to write "Frankenstein."
Galvani was working at the University of Bologna in what were then the Papal States. In 1780 he made one of those chance scientific observations that change the course of history: He saw an assistant who was operating a static-electricity generator accidently touch a steel scalpel to the sciatic nerve of a dissected frog. The frog's legs started twitching.
Of course, "chance observations" only change the world when the right person sees them. Galvani went on to show that muscle contractions would occur when an electric charge was applied to muscles or nerves. He thought his work demonstrated the existence of "animal electricity," as distinct from "natural electricity" (lightning) and "artificial electricity" (static electricity generated by friction).
Galvani reckoned that animal electricity was a fluid secreted by the brain, flowing through nerves like water through pipes. This was later shown to be wrong -- nerves are not pipes but conductors -- but his work was revolutionary. After discussions with Galvani, the physicist Alessandro Volta produced the first electric battery. Galvani's name quickly passed into language: to galvanize -- to stimulate as if by electric shock.
Mary Shelley was one of those most strongly galvanized. In 1816, after a particularly weird dream (and a challenge from poet Lord Byron), Shelley turned out a short story that would later become the novel "Frankenstein." In the introduction to the book, often called the first science-fiction novel, she wrote, "Perhaps a corpse could be reanimated. Galvanism had given token of such things."
Shelley was probably aware that after her husband's first wife, Harriet, drowned in London, doctors had attempted to revive her using electricity (they failed).
While no one has used electrical stimulation to reanimate an exhumed corpse, as the young Shelley imagined, this week in Nature researchers from the State University of New York report on a technique that is essentially derived from Galvani's work but that at first sight appears so outlandish it could be from a sci-fi comic.
Sanjiv Talwar and colleagues at the department of physiology and pharmacology wired tiny electrical probes into the brains of rats and delivered impulses to them by remote control. An operator sitting up to 500 meters away could control the movements of the rat by galvanizing electrodes implanted into the area of the brain that receives signals from the whiskers.
By stimulating this area, the rats could be steered -- made to run through a maze, up stairs, through tunnels and so on. So when the part of the brain representing the left whiskers was stimulated and, in response, the rat turned right, it was given a stimulus to the reward center. The responses were reinforced through training. Wires from the brain probes ran into a small backpack carried by each rat, containing a microprocessor and remote-controlled microstimulator.
The backpacks and electrodes worked like an ultramodern Skinner box, the device used by the American psychologist B.F. Skinner in the 1930s to condition animals to produce certain responses to an external cue (such as the sound of a bell) in order to obtain a reward (such as food). Except with brain microstimulation, the cues and the rewards are not really there -- they are virtual.
"Our system delivers cues and rewards virtually in the real world just as the Skinner system delivers them in a box," said co-author Shaohua Xu in an e-mail interview.
The scientists describe just how far they can go with their system. They can make rats jump by stimulating the medial forebrain bundle, the part of the brain concerned with reward. Rats could be controlled "purely by using video feedback from a minicamera in the backpack, without directly seeing the rat," Xu said, emphasizing that all tests were carried out within NIH guidelines for animal welfare. Rats could also be directed to do things they would instinctively avoid, like walk on the narrow top of a 2.4-meter-high fence and move around in a brightly lit, open area.
"These are manouvers that a lab rat would not routinely do," said Xu. "But, in general, the rats could not be induced to accomplish tasks that were clearly dangerous to their survival."
An ultimate aim of the research is to generate virtual cues in humans who have lost the ability to transmit sensory information to the brain, such as those with spinal injuries. But the researchers also hope to develop the electrode-backpack system to make "robo-rats" -- intelligent "robots" naturally superior to current mobile robots.
The robo-rats could be deployed in "useful real-world applications, such as search-and-rescue missions in areas of urban destruction and landmine destruction," Talwar and colleagues write in Nature.
And what of the future? Could the brain ever be stimulated indirectly, without the use of electrodes?
"We did not control the rat's mind, only guide its navigation," said Xu. "Nevertheless, in the future it will be possible to precisely target and stimulate an area within the brain. In which case a scenario akin to remote behavioral control is certainly possible -- and scary."