By DANIEL GUTIERREZCNN staffJuly 30, 2019 08:48:26A recent study published in the journal Science suggests that a neural network may be the solution to help people who suffer from autism find themselves in a virtual world.
Researchers at the University of Washington used a brain-machine system to help an autistic person navigate a virtual space.
“The task was to identify objects to place in the virtual environment,” said Dr. Daniel Gutierrez, a professor of psychology at the UW who led the research.
“We were able to use the neural network to help the person navigate in a more realistic way,” he told CNN.
The UW study used an artificial intelligence system to learn how to navigate the virtual world, based on its experience in the real world.
“If you look at the first picture you saw, it was a black-and-white screen,” said Gutierz, who presented the results at the annual meeting of the Association for Computational Neuroscience.
“We knew there were two different worlds.
It was a first step to understanding how people might be able to navigate in that one.”
Gutierrez said the results were very encouraging, adding that the researchers were hopeful that the neural system could one day be used in a classroom setting to help students with autism learn about the world around them.
“It’s a very exciting idea,” he said.
“There are a lot of opportunities for learning how to interact with this system in a way that would be useful for the classroom environment.”
I think that’s what we’re looking for in the future.
“The neural system is made of a flexible mesh of interconnected neurons that connects to the brain, and uses sensors to analyze how the brain reacts to a user’s actions.
The research team hopes that using the system in an actual classroom setting could lead to a better understanding of autism, and possibly eventually help people like it.”
Our goal is to learn about how the system works, how it works in real-time, how the user experiences it, and what we can do to improve it,” Gutierres said.
The researchers have been working with the University’s Department of Computer Science to design a prototype of the system, and are currently developing it.
The goal is for the system to be ready by 2021.