Children with Autism Spectrum Disorders Make a Fruit Salad with Probo, the Social Robot: An Interaction Study.
A social robot can spark slightly more eye-contact, but you need a full teaching plan to see real social gains.
01Research in Context
What this study did
Kids with autism made fruit salad with a friendly robot named Probo.
Researchers watched if the robot helped the kids look at people more.
They compared the robot session to a regular adult-led session.
What they found
Eye-contact went up a little when the robot was in the room.
Talking, sharing, and helping stayed the same.
The boost was small and limited to looking only.
How this fits with other research
Chiviacowsky et al. (2013) saw bigger gains: kids talked more to adults when a dinosaur robot joined.
EGranieri et al. (2020) pooled 18 trials and found tech tools can match live teaching, so the tiny eye-contact bump here is weaker than expected.
Marino et al. (2020) later turned the same idea into a full CBT program and got strong emotion-learning gains, showing the robot can work when you add lessons and repetition.
Why it matters
A robot alone is not a magic wand. Use it as a hook, then teach. Pair short robot interactions with clear social lessons and lots of practice. Track more than eye-contact—look for turns, words, and shared smiles.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start a play session with a brief robot greeting, then jump straight into scripted turn-taking drills while the child’s interest is high.
02At a glance
03Original abstract
Social robots are thought to be motivating tools in play tasks with children with autism spectrum disorders. Thirty children with autism were included using a repeated measurements design. It was investigated if the children's interaction with a human differed from the interaction with a social robot during a play task. Also, it was examined if the two conditions differed in their ability to elicit interaction with a human accompanying the child during the task. Interaction of the children with both partners did not differ apart from the eye-contact. Participants had more eye-contact with the social robot compared to the eye-contact with the human. The conditions did not differ regarding the interaction elicited with the human accompanying the child.
Journal of autism and developmental disorders, 2016 · doi:10.1007/s10803-015-2556-9