Robots can be useful as mental well-being coaches in the workplace, but the perception of their effectiveness depends largely on what kind of robot it is.
Researchers at the University of Cambridge conducted a study at a technology consultancy using two different robot wellbeing coaches, where 26 employees participated in weekly wellbeing sessions led by the robots for four weeks. Although the robots had identical voices, facial expressions, and session scripts, the physical appearance of the robots influenced how participants interacted with it.
Participants who performed their wellness exercises with a toy-like robot said they felt more connected to their “coach” than participants who worked with a humanoid robot. The researchers say the perception of robots is influenced by popular culture, where the only limit to what robots can do is the imagination. However, when encountering a robot in the real world, it often falls short of expectations.
Because the toy-like robot has a simpler appearance, participants may have had fewer expectations and ultimately found the robot easier to relate to. Participants who worked with the humanoid robot found that their expectations were not met, as the robot was unable to hold interactive conversations.
Despite the gap between expectations and reality, the researchers say their study shows that robots can be a useful tool for promoting mental well-being in the workplace. The results will be reported today (March 15) at the ACM/IEEE International Conference on Human-Robot Interaction in Stockholm.
The World Health Organization recommends that employers take measures to promote and protect mental well-being in the workplace, but the implementation of well-being practices is often limited by a lack of resources and staff. Robots have shown early promise to help address this gap, but most research on robots and well-being has been conducted in the laboratory.
“We wanted to take robots out of the lab and explore how they can be useful in the real world,” said Dr Mikol Spitale, first author of the paper.
The researchers partnered with local technology company Cambridge Consultants to develop and implement a workplace wellness program using robots. Over the course of four weeks, employees were guided through four different wellness exercises by one of two robots: either the QTRobot (QT) or the Misty II robot (Misty).
QT is a child-like humanoid robot and is approximately 90 cm tall, while Misty is a 36 cm toy-like robot. Both robots have on-screen faces that can be programmed with different facial expressions.
“We interviewed various well-being coaches, and then we programmed our robots to have a coach-like personality, highly open and conscientious,” said co-author Minya Axelson. “The robots were programmed to have the same personality, the same facial expressions, and the same voice, so the only difference between them was the physical shape of the robot.”
The robot led the experimenters through various positive psychology exercises in an office meeting room. Each session began with the robot asking participants to recall a positive experience or describe something in their life for which they were grateful, and the robot asked follow-up questions. After the sessions, participants were asked to evaluate the robot with a questionnaire and an interview. Participants did one session per week for four weeks and worked with the same robot for each session.
Participants who worked with the toy-like Misty robot reported that they had a better working relationship with the robot than participants who worked with the child-like QT robot. Participants also perceived Misty more positively overall.
“It could be that because the Misty robot is more of a toy, it meets their expectations,” Spitale said. “But because QT is more humanoid, they expected it to behave like a human, so the participants working with QT were a little overwhelmed.”
“The most common response we had from participants was that their expectations of the robot did not match reality,” said Professor Hattje Gunes from Cambridge’s Department of Computer Science and Technology, who led the study. “We programmed the robots with a script, but the participants were hoping for more interactivity. It is incredibly difficult to create a robot that can have natural conversations. New developments in large language models can be really helpful in this regard.”
“Our ideas about how robots should look or behave can hold back the application of robotics in areas where they could be useful,” Axelson said.
Although the robots used in the experiment were not as advanced as C-3PO or other fictional robots, participants still said they found the well-being exercises helpful and that they were open to the idea of talking to a robot in the future.
“The robot can serve as a physical reminder to do well-being exercises,” Gunes said. “And just saying things out loud, even to a robot, can be helpful when you’re trying to improve mental well-being.”
The team is now working to increase the responsiveness of robot coaches during coaching practice and interaction.
The research was carried out by the Engineering and Physical Sciences Research Council, part of UK Research and Innovation (UKRI). Hatice Gunesh is a member of staff at Trinity Hall, Cambridge.