You will realize that this development has great potential for good and evil. Quadriplegics may one day at least be able to feed themselves easily and do many normal functions which can be accomplished with robots. But, this opens a field of human manipulation, for what can be done by the brain can be done TO the brain as well. Think on this one.
blip, June 23 , 1999 It sounds like something out of science
fiction a rat with a small electrode sticking out of its head decides
it wants a drink and, without touching anything at all, gets a robotic arm
to bring it some water. Still, a team of neurobiologists say their rats can
control a machine with brainpower alone, and they think their technology
may someday help paralyzed people.
THE PEOPLE in the lab started calling the experiment the thinking about drinking experiment, John Chapin of Hahnemann Medical College in Philadelphia, who led the research, said in a telephone interview. But we dont know whether rats think. Whatever the rats are doing, they are controlling the robotic arm without touching anything, said Chapin, who worked with colleagues at Duke University in North Carolina. Reporting in the July issue of the journal Nature Neuroscience, they said they implanted tiny electrodes, no thicker than a hair, into the brains of six rats. It doesnt hurt the animal, Chapin said. All there is is a little plug coming out of the animals head. He runs around the cage and everything. The electrode is recording the activity of neurons on average, 46 which Chapin found was important to making the experiment work.
Earlier studies that recorded the activity of just one or a few brain cells did not work. We trained the rat initially to put his paw on a lever and to press the lever down. When the lever got pressed down, there was a robot arm that moved over to a water dropper and then brought the water back to the animals mouth, Chapin said. The rats had to control the lever carefully: If they only pushed the lever halfway, it would only bring the arm halfway to them.
Chapins team then recorded the brain activity associated with the movement of pressing the lever. We have an electronic device that converted those patterns of activity in the brain of the animal into a single electronic signal that could move the robot arm, Chapin said.
Soon they disconnected the lever from the robot arm and hooked it up to the converting device alone. They found, as other researchers have, that the brain activity controlling the movement came before the actual movement. When control of the robot arm was switched to the brain, the robot arm went over and brought water to the animals mouth before the animal even started to move, Chapin said. After a couple of days, the animals began to recognize that, and they stopped actually pressing the lever.
Chapin said if the technique can be proven safe and reliable in animals such as monkeys, which have bigger and more complex brains than rats, it might eventually be tested in people with severe paralysis. If this really becomes a workable thing, I think there are a lot of people that could use it, he said. It is important to record the signals from many neurons and not just a few, Chapin said. Of the six rats tested, he added, just four could get the arm to work. Two rats would do it a few times, and then they would stop, he said. The reason was, we were not recording enough neurons in those animals.
The robot arm would jerk around a lot and it wasnt smooth. When
the animal tried to get his mouth around it, it would kind of bop him on
the nose. They didnt like it. For complex movements, such as
those made by an artificial limb, even more neurons will be required, he
said. In principle, it should be possible to tap this information and
control a prosthetic limb, Eberhard Fetz of the University of blip
in Seattle wrote in a commentary on the findings.