Neuroscientists Demonstrate New Way to Control Prosthetic Device with Brain Signals
PASADENA, Calif.—Another milestone has been achieved in the quest to create prosthetic devices operated by brain activity. In the July 9 issue of the journal Science, California Institute of Technology neuroscientists Sam Musallam, Brian Corneil, Bradley Greger, Hans Scherberger, and Richard Andersen report on the Andersen lab's success in getting monkeys to move the cursor on a computer screen by merely thinking about a goal they would like to achieve, and assigning a value to the goal.
The research holds significant promise for neural prosthetic devices, Andersen says, because the "goal signals" from the brain will permit paralyzed patients to operate computers, robots, motorized wheelchairs—and perhaps someday even automobiles. The "value signals" complement the goal signals by allowing the paralyzed patients' preferences and motivations to be monitored continuously.
According to Musallam, the work is exciting "because it shows that a variety of thoughts can be recorded and used to control an interface between the brain and a machine."
The Andersen lab's new approach departs from earlier work on the neural control of prosthetic devices in that most previous results have relied on signals from the motor cortex of the brain used for controlling the limb. Andersen says the new study demonstrates that higher-level signals, also referred to as cognitive signals, emanating from the posterior parietal cortex and the high-level premotor cortex (both involved in higher brain functions related to movement planning), can be decoded for control of prosthetic devices.
The study involved three monkeys that were each trained to operate a computer cursor by merely "thinking about it," Andersen explains. "We have him think about positioning a cursor at a particular goal location on a computer screen, and then decode his thoughts. He thinks about reaching there, but doesn't actually reach, and if he thinks about it accurately, he's rewarded."
Combined with the goal task, the monkey is also told what reward to expect for correctly performing the task. Examples of variation in the reward are the type of juice, the size of the reward, and how often it can be given, Andersen says. The researchers are able to predict what each monkey expects to get if he thinks about the task in the correct way. The monkey's expectation of the value of the reward provides a signal that can be employed in the control of neural prosthetics.
This type of signal processing may have great value in the operation of prosthetic devices because, once the patient's goals are decoded, then the devices' computational system can perform the lower-level calculations needed to run the devices. In other words, a "smart robot" that was provided a goal signal from the brain of a patient could use this signal to trigger the calculation of trajectory signals for movement to be accomplished.
Since the brain signals are high-level and abstract, they are versatile and can be used to operate a number of devices. As for the value signals, Andersen says these might be useful in the continuous monitoring of the patients to know their preferences and moods much more effectively than currently possible.
"These signals could also be rapidly adjusted by changing parameters of the task to expedite the learning that patients must do in order to use an external device," Andersen says. "The result suggests that a large variety of cognitive signals could be interpreted, which could lead, for instance, to voice devices that operate by the patients' merely thinking about the words they want to speak."
Andersen is the Boswell Professor of Neuroscience at Caltech. Musallam and Greger are both postdoctoral fellows in biology at Caltech; Corneil is a former researcher in Andersen's lab who is now at the University of Western Ontario; and Scherberger, a former Caltech researcher, is now at the Institute of Neuroinformatics in Zurich, Switzerland.