Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder that hampers communication and social interaction. Children with ASD have difficulty sharing emotions, interests and maintaining a back-and-forth conversation. This is because they don’t completely understand body language and other non-verbal cues. It isn’t curable, but through speech and behavioural therapy, reception can be improved. Many researchers and groups have found that socially-assistive robots can be a new way of providing personalised care and help social development.
Researchers at MIT developed a sort of personalised machine learning in 2018 where the programmed robots equipped with ‘deep learning’ came to the same conclusions regarding children’s responses as human experts. The researchers used humanoid robots that are able to convey emotions by changing the colour of their eyes, limb movements, and intonation.
LuxAI, a young company in 2018, said its QTrobot can actually increase these children’s willingness to interact with human therapists, and decrease discomfort during therapy sessions. The repetitive patterns of engagement by the robots help autistic children feel at ease and focus their attention better on the presented curriculum.
More recently, a study published in Science Robotics in 2020 created a machine-learning model that, using audio and video data like dialogue and eye contact from autistic children’s interactions with the robot, can predict whether they are engaged in a given training activity and the robot’s response can automatically alter to re-engage with them. The model tested a 90% accuracy in predicting the child’s engagement despite noisy data and high variability among participants.
In 2021, a Hong Kong professor managed to develop an educational programme using role-playing robots that helps boost social skills in children with autism and helps them resolve issues such as conflicts and bullying. This programme is called Robot for Autism Behavioural Intervention (RABI) and is designed for people with ASD between ages 3 and 18. The robots engage the children in role-playing and interaction to help children recognise appropriate versus inappropriate behaviour with others.
How this is relevant for all children
Human-robot interaction (HRI) research shows that the ability of robots to talk and move have animated facial expressions, gives them better opportunity to communicate with children on a socially present and active level as compared to static screens and toys. Engaging children in learning activities is more effective with robot teaching assistants as they can be adjusted to respond to the child’s learning needs by personalizing the programmes in the parts the child is struggling with. While they can’t replace human teachers, they can offer benefits. The non-judgemental environment that a robot teaching assistant can provide can rid children of their anxiety about making mistakes in front of peers.
(With inputs from agencies)