The Nao robot looks more like a prop from a low-budget sci-fi film than the cutting edge of medical research. But a study found that children felt more comfortable confiding in the child-sized, quizzical-looking humanoid than when responding to mental health assessments with their parents, in some cases disclosing information that they had not previously shared.
The team, from the University of Cambridge, say the findings suggest a wider role for robots in assessing children’s mental health – although they said that they would not be intended as a substitute for professional mental health support.
“There are times when traditional methods aren’t able to catch mental wellbeing lapses in children, as sometimes the changes are incredibly subtle,” said Nida Itrat Abbasi, the study’s first author. “We wanted to see whether robots might be able to help with this process.”
In the study, 28 children aged eight to 13 took part in a one-to-one 45-minute session with the 60cm-tall humanoid robot, called Nao. The robot, which has a child’s voice, and, started with an ice-breaker chat and fist-bump to create a friendly atmosphere. It then asked questions about happy and sad memories over the last week, administered a questionnaire on feelings and mood and also a questionnaire used in diagnosing anxiety, panic disorder and low mood.
Children whose responses on traditional questionnaires suggested they could be experiencing mental wellbeing problems, gave more strongly negative responses when answering the same questions with the robot and some shared information that they had not disclosed when responding to in-person or online questionnaires.
Children may view the robot as a “confidant”, allowing them to divulge their true feelings and experiences, the scientists suggested. One of the parents, observing the session through a mirrored window, told the researchers they had not realised their child was struggling until hearing them respond to the robot’s questions. And previous research found that children are more likely to share private information, such as their experiences of bullying, with a robot than an adult.
“We think that when the robot is child-sized it’s easier to relate to the robot as a peer,” said Prof Hatice Gunes, who leads the Affective Intelligence and Robotics Laboratory at the Cambridge. By contrast, she said, children might respond to parents or psychologists with “what they think is expected of them rather than what they think is true”.
Gunes suggested that in future, robots could be used in schools to screen children for mental health problems, allowing children to get support at an earlier stage.
Prof Farshid Amirabdollahian, an expert in human-robotic interaction at the University of Hertfordshire, who was not involved in the work, said there was growing evidence to support the use of robots in supporting mental healthcare provision. “Children tend to show a very positive attitude to interactive technologies,” he said. “We don’t want robots to replace people but they seem to be very good tools for breaking the ice.”
The findings will be presented on Thursday at the 31st IEEE International Conference on Robot and Human Interactive Communication in Naples.
