Can a Robot’s Appearance Impact Its Effectiveness as a Workplace Wellbeing Coach?

A recent study conducted by researchers from the University of Cambridge revealed that the physical appearance of robot wellbeing coaches affects participants’ engagement with them. The research, carried out in a tech consultancy firm, involved 26 employees who took part in weekly robot-led wellbeing sessions for four weeks. Two different robot coaches were used for the experiment, both with identical voices, facial expressions, and scripts.

The employees who interacted with a toy-like robot reported feeling more connected with their ‘coach’ than those who worked with a humanoid-like robot. The researchers believe this outcome stems from the fact that people’s perception of robots is influenced by popular culture, where imagination is the only limit. Consequently, real-world robots often do not meet these expectations.

The toy-like robot’s simpler appearance might have led participants to have lower expectations, making it easier for them to connect with it. On the other hand, those who interacted with the humanoid robot found their expectations to be mismatched with reality, as the robot was not capable of having interactive conversations.

Despite the discrepancies between expectations and reality, the study demonstrated the potential of robots as a valuable tool to promote mental wellbeing in the workplace. The findings were presented at the ACM/IEEE International Conference on Human-Robot Interaction in Stockholm on March 15th.

The World Health Organization recommends that employers promote and protect mental wellbeing at work. However, the implementation of wellbeing practices is often hindered by insufficient resources and personnel. Although robots have shown early promise in addressing this gap, most studies have been conducted in laboratory settings.

Dr. Micol Spitale, the paper’s first author, explained the motivation behind the study, saying, “We wanted to take the robots out of the lab and study how they might be useful in the real world.”

The researchers partnered with local technology company Cambridge Consultants to design a workplace wellbeing program using robots. Over four weeks, employees were guided through various wellbeing exercises by one of two robots: the QTRobot (QT) or the Misty II robot (Misty).

QT, a childlike humanoid robot, stands at roughly 90cm tall, while Misty, a toy-like robot, is 36cm tall. Both robots have screen faces capable of displaying different facial expressions. The robots were programmed with a coach-like personality, featuring high openness and conscientiousness.

Participants were guided through positive psychology exercises by a robot in an office meeting room, with each session beginning with the robot prompting participants to recall a positive experience or express gratitude. The robot would then ask follow-up questions. After the sessions, participants assessed the robot through a questionnaire and an interview.

The study found that participants who worked with the toy-like Misty robot reported better working connections and more positive perceptions of the robot compared to those who worked with the child-like QT robot. Dr. Spitale suggested that Misty’s toy-like appearance could have influenced these results.

Professor Hatice Gunes, who led the research, emphasized the importance of refining robot interactivity to meet users’ expectations. She said, “It’s incredibly difficult to create a robot that’s capable of natural conversation. New developments in large language models could really be beneficial in this respect.”

Minja Axelsson, a co-author of the study, pointed out that preconceived notions of how robots should look or behave might be hindering the adoption of robotics in areas where they can be beneficial. The participants in the study, although aware that the robots were not as advanced as fictional counterparts, still found the wellbeing exercises helpful and were open to the idea of talking to a robot in the future.

The research team is now focusing on enhancing the robot coaches’ responsiveness during coaching practices and interactions.