November 9, 2013
Monkey Minds Can Move Virtual Arms
Alan McStravick for redOrbit.com - Your Universe Online
On more than one occasion, redOrbit has reported on work done at the Nicolelis laboratory at the Duke University School of Medicine. Neurobiology professor Miguel Nicolelis has pioneered the field of brain-machine interface (BMI). Previous breakthroughs have included enabling rats to “touch” infrared light and electronically connecting rat brains to create an organic computer. While these, in and of themselves, seem like really cool feats, the greater purpose of BMI is even cooler.
BMI, a field of study receiving much focus and attention in only the last 10 years is, according to the Walk Again Project website, “…one of the most exciting – and promising – areas of basic and applied research in modern neuroscience.”
The Walk Again Project, led by the Duke Center for Neuroengineering, is a multinational collaborative effort to develop high-performance brain-controlled prosthetic devices meant to enable paralyzed victims to regain their mobility. The Brazilian research funding agency FINEP awarded a research grant of $20 million recently, and with the latest news out of the Nicolelis lab, they are wasting no time in using that funding to further advance the field of BMI from science fiction to science fact.
This week, it was announced the end goal is getting even closer. This is because Nicolelis and his team were able to successfully allow rhesus monkeys to move two arms of a virtual primate on a screen using only their minds. In what is being hailed as a significant first, the two test monkeys have been the only animals able to control both limbs thanks to electrodes surgically implanted in their brains.
According to Nicolelis, who, as reported previously at redOrbit, has set a goal of having a paralyzed person outfitted in his neurally controlled exoskeleton deliver a kick at next year’s World Cup finals in Brazil, this latest success represents an important milestone in the future development of exoskeletons that can be operated as though they were a natural extension of the wearer’s own body.
From typing on a keyboard to opening a can, life is full of movements that require two arms, Nicolelis said in a statement. “Future brain-machine interfaces aimed at restoring mobility in humans will have to incorporate multiple limbs to greatly benefit severely paralyzed patients.”
Even though the “virtual” monkeys lacked a natural authenticity, the test monkeys quickly grasped the concept of how to make the arms of the avatar move, reach for and hold targets that blinked and moved on the screen.
In an interview with Nidhi Subbaraman of NBC News, Nicolelis commented that the monkeys reacted “as if it were them in the virtual space.” If the monkey was able to manipulate the virtual hand over a target for more than 100 milliseconds, they received, as a reward, a sip of fruit juice.
Other researchers in the field have thus far been unable to replicate the results as their monkeys have only been able to move a robotic arm, or only a single virtual arm or cursor with their minds.
The study, published this week in the journal Science Translational Medicine, reports both monkey subjects were able to move both of the virtual arms in four different directions. According to Nicolelis, his experiments have shown that a record number of neurons simultaneously recorded in non-human primates – almost 500 in one monkey – were achieved.
In the process of the research, the team trained the monkeys to control the virtual arms, first with joysticks, and then later, using only the power of their minds. What was perhaps most interesting was that the monkeys were able to, over time, improve their ability to manipulate the arms. The team points out that the monkeys' brains exhibited remarkable plasticity in the regions of the brain being recorded. Furthermore, the study claims the monkeys were incorporating the idea of the virtual arms into their internal body image.
According to Nicolelis, the results of their research suggest ensembles of neurons, rather than individual neurons, form the basic unit of normal movement control. Trying to paint with a broad brush in determining neural activity associated with right or left arm movements alone is insufficient in predicting the brain activity involved in moving both arms in concert.
This is, by no means, the only research being conducted in the field of BMI at this time.
Also in the NBC News story were examples of researchers out of Stanford University and the University of Pittsburgh.
A 2012 study by UPitt neurobiologist Andrew Schwartz presented a system that allowed a 58-year-old woman, who had been paralyzed since her mid-40s, to control a robotic arm well enough to bring her a drink of coffee. The year before, the same research team enabled a paralyzed man to high-five his girlfriend using a neurally controlled robotic arm.
Other researchers are exploring options that would not require an exoskeleton, necessarily. Work is being done in the hopes they can one day re-invigorate a patient’s existing limbs by connecting the electrodes in the brain directly to muscle tissue, basically doing and end-run around the function of the injured spine.
Finishing his interview with NBC News, Nicolelis exhibited a high level of confidence in his World Cup goal. “It’s going – it’s on schedule,” he said.