Electric Thinking Caps Aid Learning
Surgical robots are in high demand in the operating room. But in spite of their ability to assist surgeons perform high-precision operations, the number of hours they can be used is limited due to being expensive and full of moving parts with finite lives. Because of that, surgeons learning to operate a machine like a Da Vinci robotic surgical system to do their training in a VR setting before trying out their skills on the real thing. But however many repetitions of a task a surgeon might enact in the virtual setting, the new skills may not fully translate to the real world.
How do you get surgeons to better hold onto their VR-learned abilities? A team of researchers at Johns Hopkins has the answer: Zap their brains.
That’s just what Jeremy D. Brown, a professor in the department of mechanical engineering at the university, and his colleagues, did to the subjects in a recent experiment. Participants trained to drive a curved needle through a defined trajectory. One group trained in VR simulation—trying to thread the needle 40 times—and then repeated the task on a real-world robot. Another group trained in the real-world first and then tried to repeat the task in virtual reality.
Experiments conducted by mechanical engineering professor Jeremy Brown trained subjects on the task of threading a needle using robotic grippers. Image: Johns Hopkins University
Each of those groups was divided, with some doing their learning with a small current applied to the back of the head and some without the electric stimulation.
With or without the transcranial stimulation, all the groups got better during their initial training sessions. But when they switched environments, the value of the electrical stimulation became shockingly apparent.
“When we ask them to do the task in the real world, the errors actually got significantly worse,” Brown said of the group that trained in the VR setup without the current.
But the participants that trained in the virtual world while being stimulated held on to their new skills. “Instead of the errors getting worse when they moved to the real-world platform, their errors basically did not significantly change,” Brown said. “The simulation allowed them to transfer their skill from VR to the real-world better than the group that did not get stimulated.”
Why does the electric stimulation help this transference of learning from the virtual world to the real one?
“I’ll be honest, I don't fully know, but I do have my suspicions,” Brown said.
One of those suspicions has to do with the cerebellum, a small section of brain at the back head that received the stimulation in the experiment. It’s responsible for what Brown calls “ballistic movements,” or actions that are done before any feedback channels in the brain kick in.
“What we think is happening is, when you stimulate the cerebellum and make it highly excitable, you are actually increasing someone’s ability to build a good feed-forward model,” he said.
More like this: Video Games Offer Brain-Computer Interface Training Ground
Another possibility is that there’s just something about the virtual world that lets the brain—or the person—take whatever happens there less seriously. And, somehow, the added electricity applies some kind of “truth” label to an experience.
“Think about playing video games: dying in a video game is this nonconsequential thing. It doesn't hold the same weight, or the gravity, of dying in the real world,” Brown said. “Maybe when you operate in VR, the brain is not in this sort of prime state where it understands that these things are consequential in some significant way. And doing the brain stimulation primes the brain to say ‘No, no, no, no, this is actually very important.’ ”
There’s much work to be done before such training becomes standard. For one thing, as of now, no one knows if the skills learned in VR with cranial stimulation stick around and can be called up a day later, a month later, or a year later. Brown would also like to repeat the experiment with the kind of skills that surgeons must learn for real surgery.
Editor's choice: AI Helps Soldiers Bark Orders Telepathically to Robot Dogs
“I would love to do this on an actual suturing task,” Brown said.
He also hopes to run a similar experiment, but with EEG to better figure out what’s actually going on in the brain during this cranial stimulated virtual learning.
However it works, the find is likely to open up worlds of more effective training, and will may help more than just surgeons-in-training learning complex moves. Soldiers in virtual combat zones could benefit from the cranial stimulation, as could athletes.
It may also be possible that adding current to heads in non-virtual settings could have a similar effect. Could musicians practicing at home, for example, better transfer their efforts to the stage if they receive a little transcranial stimulation while learning their parts?
“I think yes,” Brown said. “The stimulation could lead to better performance in other tasks outside of the surgery domain.”
Michael Abrams is a technology writer in Westfield, N.J.
How do you get surgeons to better hold onto their VR-learned abilities? A team of researchers at Johns Hopkins has the answer: Zap their brains.
That’s just what Jeremy D. Brown, a professor in the department of mechanical engineering at the university, and his colleagues, did to the subjects in a recent experiment. Participants trained to drive a curved needle through a defined trajectory. One group trained in VR simulation—trying to thread the needle 40 times—and then repeated the task on a real-world robot. Another group trained in the real-world first and then tried to repeat the task in virtual reality.
Experiments conducted by mechanical engineering professor Jeremy Brown trained subjects on the task of threading a needle using robotic grippers. Image: Johns Hopkins University
With or without the transcranial stimulation, all the groups got better during their initial training sessions. But when they switched environments, the value of the electrical stimulation became shockingly apparent.
“When we ask them to do the task in the real world, the errors actually got significantly worse,” Brown said of the group that trained in the VR setup without the current.
But the participants that trained in the virtual world while being stimulated held on to their new skills. “Instead of the errors getting worse when they moved to the real-world platform, their errors basically did not significantly change,” Brown said. “The simulation allowed them to transfer their skill from VR to the real-world better than the group that did not get stimulated.”
Why does the electric stimulation help this transference of learning from the virtual world to the real one?
“I’ll be honest, I don't fully know, but I do have my suspicions,” Brown said.
One of those suspicions has to do with the cerebellum, a small section of brain at the back head that received the stimulation in the experiment. It’s responsible for what Brown calls “ballistic movements,” or actions that are done before any feedback channels in the brain kick in.
“What we think is happening is, when you stimulate the cerebellum and make it highly excitable, you are actually increasing someone’s ability to build a good feed-forward model,” he said.
More like this: Video Games Offer Brain-Computer Interface Training Ground
Another possibility is that there’s just something about the virtual world that lets the brain—or the person—take whatever happens there less seriously. And, somehow, the added electricity applies some kind of “truth” label to an experience.
“Think about playing video games: dying in a video game is this nonconsequential thing. It doesn't hold the same weight, or the gravity, of dying in the real world,” Brown said. “Maybe when you operate in VR, the brain is not in this sort of prime state where it understands that these things are consequential in some significant way. And doing the brain stimulation primes the brain to say ‘No, no, no, no, this is actually very important.’ ”
There’s much work to be done before such training becomes standard. For one thing, as of now, no one knows if the skills learned in VR with cranial stimulation stick around and can be called up a day later, a month later, or a year later. Brown would also like to repeat the experiment with the kind of skills that surgeons must learn for real surgery.
Editor's choice: AI Helps Soldiers Bark Orders Telepathically to Robot Dogs
“I would love to do this on an actual suturing task,” Brown said.
He also hopes to run a similar experiment, but with EEG to better figure out what’s actually going on in the brain during this cranial stimulated virtual learning.
However it works, the find is likely to open up worlds of more effective training, and will may help more than just surgeons-in-training learning complex moves. Soldiers in virtual combat zones could benefit from the cranial stimulation, as could athletes.
It may also be possible that adding current to heads in non-virtual settings could have a similar effect. Could musicians practicing at home, for example, better transfer their efforts to the stage if they receive a little transcranial stimulation while learning their parts?
“I think yes,” Brown said. “The stimulation could lead to better performance in other tasks outside of the surgery domain.”
Michael Abrams is a technology writer in Westfield, N.J.
Technology is rapidly changing the world we live in. Explore content covering the latest trends in autonomous technology including artificial intelligence, self-driving cars, and automation.