- Video: Robot helps researchers crack autism code
- More: Most Popular stories, videos and more
- More: Get breaking news alerts
"He would just turn and face the wall," said Tonya Mirtes, a mother with an autistic son.
Somehow a robot masquerading as a basketball game helped Daniel Mirtes come out of his shell.
"Just with this robot, it was able to keep him engaged," said Tonya.
"You shoot hoops in it. If I'm stressed out then it will slow down," said Daniel.
Sixteen-year-old Daniel has autism. It looks like he's playing basketball, but he's actually going several rounds with a robot.
It's helping researchers crack the autism code. Daniel's heart rate, skin temperature and muscle movements are recorded. From that, the robot reads and responds to his mood.
"If the robot determines that the child is getting stressed out, the robot will change the speed of the game and may play relaxing music," said Nilanjan Sarkar, Ph.D., from Vanderbilt University's departments of mechanical and computer engineering.
"The amazing thing is you can put it all together and learn what a child is feeling," said Wendy Stone, Ph.D., from Vanderbilt University.
The robot helps humans understand what causes anxiety and calmness -- something autistic kids can't put into words.
"If we can adapt our behavior, we're going to have much more success," said Stone.
Researchers found robots predicted the child's emotional state correctly more than 80 percent of the time. The robot was about as good at identifying a child's feelings as an experienced therapist. Robots are helping humans understand autism.
Researchers believe the robots work well for kids with autism because they are more predictable and consistent than humans -- something these children can relate to and feel comfortable with. They're working on creating a smaller version of the robot, so parents can work with it at home.
Web Extra Information: Robots Unravel Autism
Autism is a brain development disorder defined by impaired social interaction and communication, and restrictive and repetitive behavior. Many of these signs present themselves before a child's third birthday. Autism impacts many parts of the brain, but how it affects the brain is still not clearly understood. The autism spectrum disorders (ASD) also include the related conditions Asperger syndrome and PDD-NOS, which have fewer signs and symptoms. Controversies surround proposed environmental causes of autism.
The number of people diagnosed with autism has increased dramatically since the 1980s, but the question of whether actual prevalence has increased is still debated. It's estimated one baby in every 150 born today in the United States will be diagnosed with an ASD, making them more common than pediatric cancer, diabetes and AIDS combined. Treatment involves a combination of behavioral, educational, physical and speech therapies, which are sometimes combined with medication.
It's estimated that autism currently costs the U.S. more than $90 billion per year, and according to Vanderbilt University, that cost is projected to double by 2017.
ROBOTS & AUTISM:
A lot of current research is focused on robots working with autistic children. So far, results show children are attracted to robots, raising the promise that appropriately-designed robots could play an important role in development. Researchers at Vanderbilt University Medical Center built a robot that can monitor the emotional state of autistic children and respond to their reactions.
Nilanjan Sarkar, Ph.D., from Vanderbilt University's departments of mechanical and computer engineering, developed a method that uses physiological measurements, including heart rate, skin response, temperature and muscle movement, to monitor the emotional state of these children.
Children ranging in age from 13 to 16 who had been diagnosed with an ASD were connected to a variety of sensors. Then, the children were asked to play two games. One was the computer game Pong. The other was a variant of Nerf basketball with the hoop and backboard attached to the end of a robotic arm that moves it back and forth or up and down.
Researchers collected the data and built a model for each child that predicts his or her emotional state of anxiety, happiness or engagement with an accuracy of more than 80 percent. That means the robot can read the physiological cues of the child playing the game and then response to the child's needs. Researchers believe because robots are consistent and predictable, they are especially helpful when working with autistic children.