Robots can improve mental well-being in the workplace, but they need to look right, says research.
The perception of how effective the machines are depends to a large extent on what the robot looks like, according to the study.
Researchers from the University of Cambridge conducted a study at a technology consulting firm using two different robotic well-being coaches.
26 employees participated in weekly, robot-led wellbeing sessions for four weeks.
While the robots had identical voices, facial expressions, and scripts for the sessions, their physical appearance affected how people interacted with them.
Those who did their wellness exercises with a toy-like robot said they felt more connected to their “coach” than people who worked with a humanoid robot.
The researchers say perceptions of robots are influenced by popular culture, where the only limit to what robots can do is imagination.
But when you face a robot in the real world, it often doesn’t live up to expectations.
According to the researchers, because the toy-like robot looks simpler, the researchers may have had lower expectations and eventually found it easier to talk to and bond with the robot.
Those who worked with the humanoid robot found their expectations didn’t match reality because the robot wasn’t able to hold interactive conversations, they said.
The researchers worked with local technology company Cambridge Consultants to design and implement a program for workplace wellbeing using robots.
Over the course of four weeks, employees were guided through four different wellbeing exercises by one of two robots: either the QTRobot (QT) or the Misty II robot (Misty).
The QT is a childlike humanoid robot and stands around 90cm tall, while Misty is a 36cm tall toy-like robot.
Both have screen faces that can be programmed with different facial expressions.
dr Micol Spital, the paper’s first author, said: “It could be that, being more toy-like, the Misty robot met their expectations.
“But since QT is more humanoid, they expected it to behave like a human, so participants who worked with QT may have been a little underwhelmed.”
After interviewing various wellness coaches, the researchers programmed the robots to have coach-like personalities with high levels of openness and conscientiousness.
Professor Hatice Gunes of the Cambridge Department of Computer Science and Technology, who led the research, said: “The most common response from participants was that their expectations of the robot did not match reality.
“We programmed the robots with a script, but the participants were hoping for more interactivity.
“It’s incredibly difficult to design a robot capable of natural conversation. New developments in large language models could be really beneficial in this regard.”
Co-author Minja Axelsson said: “Our ideas of how robots should look or behave could inhibit the uptake of robotics in areas where they can be useful.”
The results will be presented at the ACM/IEEE International Conference on Human-Robot Interaction in Stockholm.