Good Vibrations: Passive Haptic Learning Could Be a Key to Rehabilitation

Caitlyn Seim - PHL

It has shown positive results in spinal injury recovery. Can it do the same for stroke patients?

It was around 2008, and Kevin Huang was a master’s student at Georgia Tech under the advisement of School of Interactive Computing (IC) Professor Thad Starner. Huang, an intrepid student, had big ideas within the field of haptics and wanted to pursue the creation of a full-bodied exoskeleton.

“Of course, that costs about a million dollars,” Starner, an associate professor at the time, said with a laugh. “So, I said, ‘Here’s an idea. Why don’t you try making this glove where you put vibrating motors above each finger and see if you can teach people how to play piano passively?’”

Starner was suggesting a new approach that later became known as passive haptic learning. Users would wear the gloves while doing other activities, like reading email or driving, and have the device tap each finger in the appropriate sequence for a piano melody over and over again. The hope was that the repetition would give the wearer the motor memory to later take off the gloves and play the song on the piano, perhaps even a song they had never heard before.

Huang liked the idea and ran with it. Starner, of course, thought the idea had merit or never would have suggested it. But even he was surprised with Huang’s results. The glove worked. Users, as was later demonstrated during a live segment on CNN, could quickly learn a tune in a short span of time.

“I don’t believe this,” said Starner, recalling his initial reaction. “I had never heard of anything like this before in my life. We went and did a 16-person user study, and it came back with even better results. I figured we had something real there, and it was time to dig a little deeper.”

What Starner didn’t know was that the initial whim was laying the foundation for an entirely new field of study, a field that has demonstrated positive results in sensation-based learning for music, Morse code, and even Braille.

VIDEO: These Gloves Can Teach You How to Play Piano. And Maybe Heal Your Brain (Freethink's Superhuman)

Current IC Ph.D. student Caitlyn Seim and her team of researchers has taken this initial discovery to the next level. She defined this new field of study when she unlocked even more important secrets within the real-world impact of passive haptic learning. If it can improve dexterity in healthy hands, could it do the same for those with limitations? Could it, in fact, lead to rehabilitation for someone who has suffered a traumatic spinal injury? Could it even aid in recovery from a stroke?

The elephant in the room

Seim was an undergraduate in electrical engineering at Georgia Tech in 2013 around the time Starner and former Ph.D. student Tanya Estes were beginning to understand some of the ramifications of their method. She had done some work with IC Professors Gregory Abowd and Jim Rehg, and Starner approached her about working with him in this field of passive haptic learning.

Estes and Starner had recently partnered with the Shepherd Center for spinal cord and brain injury rehabilitation to test whether the seemingly random stimulation of the fingers, as demonstrated in the piano project, could lead to increased sensation dexterity in the hands for injury patients.

It was a logical next step in the research, Starner said, and the results were encouraging. In the study, participants who were injured more than a year prior wore the Mobile Music Touch glove that led to learned skills on a piano. They participated in simple piano lessons and evaluations indicated stastistically significant improvement among the experimental group.

They were beginning to scratch the surface of what passive haptic rehabilitation was able to achieve. But, Seim said, there has always been one elephant in the room for anyone in the rehab space.

“Stroke,” Seim said. “It’s the clear elephant in the room. It’s the No. 1 cause of long-term disability in the United States and a leading cause globally.”

Not only is it a huge financial burden, patients also have precious few options for recovery. Exercise-based therapy such as constraint-induced movement is the state of the art. For immobile hands, Botox injections are also common.

“But that’s only temporary,” Starner said. “It’s not retraining the body. It’s for relief, not getting your hand back.”

The exercise-based therapy can help, but it is painful, expensive, and only available to about 50 percent of patients who meet the baseline dexterity level to begin treatment. The other half have been rendered too disabled to be eligible, so compensatory strategies like spousal assistance are encouraged.

A new option?

With the positive results in the partial spinal cord injuries study, the thought was that perhaps this stimulation-based method could have a similarly positive impact for stroke patients, as well. Like in the previous study, patients wear the gloves, this time for three hours each day for two months.

Seim, who Starner said recruited her own subjects via news groups and mailing lists, takes measurements weekly. Volunteer clinicians also take measurements at the beginning, middle, and end of the cycle.

While the team is currently preparing a publication on their work and is not yet ready to release results, Seim said the findings have been encouraging.

“It’s exciting,” Seim said of the study. “We aren’t looking for complete recovery but if we can actually get patients to regain control of their fingers or hands, they can do so much more for themselves.”

This method has already shown plenty of promise. If it impacts stroke recovery, that’s already a significant portion of the population. And perhaps passive haptic learning is a key that could unlike even more avenues of study.

“I think there’s just enormous potential in the area of haptics, wearables, and tapping into this area of passive stimulation,” Seim said. “As I finish my Ph.D. and see the research landscape, I see that we are uncovering a new paradigm beyond just the awesome applications like learning a melody on the piano or having a great system to teach Braille. These are great applications in themselves, but this is a whole new cognitive approach, and it’s very exciting.”

Contact: 

David Mitchell

Communications Officer

david.mitchell@cc.gatech.edu