Can Humans Trust Robots More if They Appear More Human-Like?

A recent study has delved into the intriguing question of whether humans can trust robots more if they appear more human-like. The findings provide insights into the dynamics of human-robot collaboration and challenge conventional assumptions.

In this research, scientists explored human trust in robots, mainly focusing on the role of eye gaze in shaping this trust. Human beings are inherently social creatures, and eye contact plays a crucial role in social interactions, conveying intentions and allowing them to adapt to the intentions of others.

Robots' Eyes Are the Windows to the Soul?

The question at the heart of this study was whether the presence of eyes on robots, mimicking human-like gaze, influences the level of trust humans place in these machines when collaborating with them.

The study centered on "cobots," collaborative robots designed to assist humans in various tasks, ranging from industrial to medical applications. Some of these cobots are equipped with eyes and designed to resemble humans to a certain extent, with the belief that this anthropomorphic appearance might ensure the naturalness of their interactions with humans.

To investigate this, Artur Pilacinski and his team of researchers at the Ruhr University Bochum in Germany, the University of Coimbra, and the University of Trás-os-Montes and Alto Douro, both in Portugal conducted experiments with 38 participants aged between 18 and 42.

They both employed subjective and objective measures, including heart rate, pupil size, and task completion time, to assess human trust levels when collaborating with eyed and non-eyed robots of the same type.

The results, outlined in the study titled "The robot eyes don't have it. The presence of eyes on collaborative robots yields marginally higher user trust but lower performance," challenge conventional assumptions.

Although participants expressed slightly greater trust in robots fitted with eyes, their performance revealed a contrasting narrative. When collaborating with robots lacking eyes, participants demonstrated enhanced task efficiency and exhibited larger pupil sizes, potentially reflecting heightened engagement with the task.

These results imply that humans may not inherently demand human-like attributes in robots to establish trust and productive collaboration. Instead, the data suggest that individuals may find it more comfortable to cooperate with robots with machine-like characteristics that do not mimic human eyes.

Human-Like Eyes on Robots

In a statement, Artur Pilacinski noted that "the cobot's gaze may not be that important for manual collaboration." The objective metrics, such as task completion time and pupil responses, indicate that cooperation with non-eyed robots may be more comfortable.

This research aligns with emerging suggestions that anthropomorphism, or the tendency to attribute human characteristics to non-human entities like robots, may not always be beneficial for collaborative robots.

The study highlights the complex interplay between objective and subjective markers of trust in human interactions with artificial agents. In essence, the presence of human-like eyes on robots may marginally increase subjective trust, but it can come at the cost of reduced task performance and comfort in collaboration.

This study encourages a reevaluation of the role of anthropomorphism in the design and deployment of collaborative robots, emphasizing the importance of objective measures in assessing human-robot interactions.

"We found that while collaboration with eyed cobots resulted in slightly higher subjective trust ratings, the objective markers such as pupil size and task completion time indicated it was in fact less comfortable to collaborate with eyed robots," the researchers wrote.

The findings of the team were published in the journal Heliyon.