Mind-controlled machines have been the subject of intense research in recent years, with a focus on improving the lives of people with disabilities. At this fascinating juncture between neuroscience and robotics we have seen experimental brain-machine interfaces (BMIs) used to fly drones, control telepresence robots and even switch channels on a TV. Now scientists at Duke University have developed a similar system that enables monkeys to drive a wheelchair using nothing other than brainwaves.
Certain research groups have found success with BMIs using electroencephalogram (EEG) caps, which are noninvasive headsets fitted with electrodes that detect brainwaves through the scalp and translate them into control commands. But in the view of the Duke scientists, this approach has its limitations.
“In some severely disabled people, even blinking is not possible,” says Miguel Nicolelis, senior author of the research paper. “For them, using a wheelchair or device controlled by noninvasive measures like an EEG, may not be sufficient. We show clearly that if you have intracranial implants, you get better control of a wheelchair than with noninvasive devices.”
These implants Nicolelis refers to number in the hundreds, and consist of hair-thin microfilaments planted in the premotor and somatosensory regions of two monkey brains. But it wasn’t a matter of simply feeding them in and watching the animals effortlessly scoot around the lab.
Beginning in 2012, the scientists trained the animals by wheeling them towards a bowl of grapes and recording their electrical brain activity. They programmed software to convert these brain signals into digital commands that dictated control of the wheelchair. Eventually, as the monkeys thought about moving towards the grapes, the system translated those thoughts into actual operation of the wheelchair and moved them toward their goal. They then became faster and more efficient at retrieving the grapes over time.
The team also began to observe brain signals indicating that the monkeys were contemplating the distance from the wheelchair to the the bowl of grapes, something they had not expected.
“This was not a signal that was present in the beginning of the training, but something that emerged as an effect of the monkeys becoming proficient in this task,” says Nicolelis. “This was a surprise. It demonstrates the brain’s enormous flexibility to assimilate a device, in this case a wheelchair, and that device’s spatial relationships to the surrounding world.”
In this particular trial, the scientists measured the activity of almost 300 neurons in the two monkeys, though they say the technology has previously allowed the recording of up to 2,000. They are now working to record more neuronal signals to attain better accuracy with the monkeys, with a view to progressing to trials in humans.