Do robots understand children better than their own parents? The latest research suggests that they might. A recent study from the University of Cambridge has found that robots can identify children’s mental wellbeing concerns better than both parent-report and self-report assessment methods.

Previously, parent-report and self-report tests were thought to be largely effective in understanding the prevalence of mental health issues in children. However, there are multiple reasons why such methods may be problematic, including the reluctance of children to disclose sensitive information about their mental wellbeing to an unfamiliar adult.  

Robots have already been demonstrated as useful tools in encouraging children to divulge sensitive and private information involving bullying, violence and abuse. Such successes have inspired a team of Cambridge roboticists, computer scientists and psychiatrists to investigate how robots could be used to assess children’s mental wellbeing.

The researchers used a child-sized humanoid robot to assess the mental wellbeing of 28 children aged between 8 and 13 years old. Each child took part in a one-to-one 45-minute session with a robot called Nao, which is equipped with innovative features, including human-like movement and speech production. The children interacted with the robot via both verbal communication and physical contact with sensors on the robot’s hands and feet.

‘Children might see the robot as a confidante – they feel like they won’t get into trouble if they share secrets with it.’

The robot was able to assess the children’s mental health using a series of questionnaires and interactive tests. The researchers found that the robots showed greater sensitivity in detecting children with mental wellbeing concerns compared to the traditional parent-report and self-report assessment methods.

This study comes in response to the societal changes during the COVID-19 pandemic – including social isolation, study-from-home arrangements and limited finances – which severely impacted the mental wellbeing of children. Recognition of such an upsurge (which had begun even prior to the pandemic era) in the incidence of mental wellbeing concerns in children has led to a monumental call to action by both the healthcare and psychological support fields, particularly due to both the limited resources and immense challenges in understanding and addressing the unique welfare needs of children.

First author Nida Itrat Abbasi commented: “Since the robot we use is child-sized, and completely non-threatening, children might see the robot as a confidante – they feel like they won’t get into trouble if they share secrets with it.”  

However, whilst this novel research demonstrates that robots can effectively identify children with mental wellbeing related problems, co-author Dr. Micol Spitale stressed: “We don’t have any intention of replacing psychologists or other mental health professionals with robots, since their expertise far surpasses anything a robot can do.”

Going forward, the researchers are keen to diversify their investigation to consider important non-verbal aspects of the child-robot interaction, such as physiology and body gestures.