What the latest research tells us about robots for autism.
One of the questions we get asked about Milo and the other robots at RoboKind is why use a robot to help teach autistic students instead of more one-on-one time with teachers and parents?
We believe so strongly that assistive robots – like Milo – can help to teach social-emotional skills to autistic kids that we’ve built our entire company on the premise.
But why do these robots work to improve student engagement and learning- often better than a teacher or therapist alone? And what is it about robot-assisted autism therapy that makes it effective?
Here’s everything you need to know about the use of robots in autism therapy.
How It Works
Preliminary studies show that robot-assisted autism therapy (RAAT) may be helpful in a variety of ways.
For one thing, autistic students often engage more with an assistive robot than they do with the human adults who work with them. A study through the University of Texas at Dallas showed that students interacted with Milo 87.5% of their time with it and only 3-4% of the time they spent with the therapist alone.
The reasons for this disparity are multi-faceted, varying among individual children. However, one definite benefit is that Milo gives simple and consistent responses that are easy to understand – leaving behind complex social cues that tend to bog down human interactions.
Because the behaviors of robots are generally more predictable than that of humans, it's easier for autistic students to interact with them – and ultimately learn from them.
RAAT can also facilitate an autistic student’s interactions with the adults who work with them, like behavioral therapists, special education teachers, speech and language pathologists. When the student socially shares interactions with the robot with another adult, it gradually allows a greater degree of expression.
By increasing engagement with the learning environment, robots have the potential to remove many of the SEL barriers that exist for autistic kids.
The bonus for the programs using robots in the curriculum is the extensive data collection and achievement tracking available. For education professionals, having a goldmine of data to create and track IEP goals and inform individualized learning opportunities is priceless.
Of course, the ability to interact with a robot means nothing unless the treatment is effective.
So does it work?
What the Research Shows
The development of robot-assisted autism therapy is in its early stages, and there’s still a lot that we don’t know. However, some early studies show promise.
Recently, a team of researchers at the University of California found that the robot they used was 90% accurate in predicting the engagement of autistic children. This remarkable accuracy has hopeful implications for crafting an engaging educational plan for individual students.
The same team of researchers also found that consistent sessions with a robot for autism improved students’ level of empathy towards peers, and made them more engaged in relationships with family members, as well.
Other studies have yielded similar results, showing that robot-assisted autism therapy improves communication skills and encourages children to engage in symbolic play.
Of course, it remains difficult to gauge the true efficacy of robots for autism until a greater body of research is available. Individual differences, such as whether or not students were verbal (or pre-verbal), and the severity of their symptoms, seemed to impact the results. But that understanding helps us to make Milo and his friends that much more powerful.
Implications For Use
As we gather more and more research on the effectiveness of robots like Milo and his friends, we see so many positive results that are specific to the child’s engagement with the robot. But much of that engagement happens with an adult present.
Using RAAT does not replace valuable one-on-one time with parents and trained specialists. Any kind of therapy for autistic children only works as well as those who administer it.
An autistic student (of any age) is likely to benefit from robot-assisted autism therapy if:
They do not have access to at least 20 hours of therapy per week.
Therapy has felt like a “one-size-fits-all” approach, rather than individualized to their needs.
They need targeted instruction in building social skills.
They are pre-verbal and/or high functioning (these students had the best results in most studies).
When utilized properly, a robot for autism (like Milo) can help remove student barriers to social interaction and give them skills that they can then generalize away from the robot.
In fact, it may make that individual time with parents and teachers all the more productive and meaningful.