Using Movement to Diagnose and Treat Autism

  • Share
  • Read Later
Kim Gunkel / Getty Images

By taking advantage of disrupted motor connections in the brains of autistic children, researchers say it may be possible for affected kids to guide their own therapy.

Scientists from Rutgers University and Indiana University have developed a novel and objective way to both diagnose and treat children with autism, by focusing on the unique ways that autistic brains process movement.

Elizabeth Torres, a computational neuroscientist at Rutgers University, explores how movements reflect the way people interact with and sense their environments. Patterns in these movements can reflect brain processes and connections, and that could be helpful in understanding autism.

“The way that we study the brain is quite disembodied. We pay attention to the central nervous system—brain and spinal chord–and we don’t pay attention to the peripheral nervous system,” says Torres, referring to the network of nerves involved in relaying sensory information such as touch, sight and smell. “This plays a pretty important function in self-regulation and autonomy, and it is not often considered in autism and in much of brain research.”

Movement can influences our perception of the world around us, and our ability to sense the environment can also change our movements. “Movement is a form of sensory input that travels back to the brain as a form of feedback, continuously,” she says.

(MORE: Unique Brain Pattern Could Predict Autism in Youngest Children)

The central nervous system constantly receives and processes this feedback in order to produce the appropriate actions. During normal development, this system learns to anticipate sensory consequences, like how a baby learns to suction its mouth for feeding. But this process may not mature in the same way in autistic children, the researchers discovered.

In two papers published in the journal Frontiers in Neuroscience, Torres, working with a computer scientist and physicist, described a way to both diagnose autism via movement patterns, and potentially treat the condition using similar action-based strategies. They developed a method that focuses on the spontaneous movements that autistic children, even infants, make unintentionally. The research team measured tiny fluctuations of movement among autistic patients, and compared these movements to those of normally developing subjects.

This strategy was able to diagnose autism among children aged three to 25, but even more exciting was the fact that the movement profiles were unique enough to distinguish how severely affected children were by the developmental disorder. All the autistic participants — regardless of their age — were essentially stunted in their ability to process movement by age three. By age four, these patterns in normally developed youngsters should be predictive and reliable. By college age, they are highly predictive, and adults can anticipate how their actions impact their environment and vice versa. But kids with autism are not successfully forming these connections.

“Their feedback is nothing. It’s corrupted and noisy and random. It is very unlikely that they can build anchors or a frame of reference to understand the physical world around them because everything that we process has to become part of the person, and for that to make sense to the person, it has to be integrated through movement,” says Torres. This could explain the difficulties that autistic children have with social interactions, since visual movement is also important for interpreting and responding to social cues, many of which involve movements.

(MORE: New Study Suggests Autism Can be ‘Outgrown’)

To test the validity of their action-based platform, the researchers then devised a way to translate their findings into individualized treatments. They created a Wii-like digital workspace with a computer screen loaded with short snippets of cartoons, videos or television shows that the children could choose and activate by waving a hand that was attached to a sensor. The children “learned” which actions opened which onscreen activities, and eventually were able to open just the ones they liked. The process likely helped them to build the feedback connection between their movement and intention that had not developed properly.

(MORE: Autism and Air Pollution: The Link Grows Stronger)

Watch a video of the therapy here:


The kids were able to both learn the movements and remember them for several weeks until the next session. By the end of the study, they were better able to control their bodies and understand their environment.

Whether the approach will become a reliable way of diagnosing and treating autism, particularly for infants, will require more testing, but introducing movement as another modifiable characteristic of autism could help more children with the disorder to connect with their surroundings and ultimately interact with them more successfully.