A Robot’s Gaze- Do robots have to be human-like for us to trust them?
This post is also available in: 
עברית (Hebrew)
New research states that humans might not need robots to seem human-like in order to trust and work with them, and even seem to collaborate better with machine-like robots.
We as humans are a collaborative species, therefore the eyes are a critical element for our social life. The question arose as to whether “eyes” on machines help us collaborate with robots.
The name “Cobots” was given to collaborative robots created to help humans in various fields, and some of them were designed with added eyes for a more anthropomorphic appearance to enhance the comfort levels of humans working with them.
According to Techxplore, Artur Pilacinski and his team used 38 participants aged between 18 and 42 and analyzed subjective and objective measures (like heart rate, pupil size, and task completion time) to assess the level of human trust when collaborating with eyed and non-eyed robots of the same type.
The article “The robot eyes don’t have it. The presence of eyes on collaborative robots yields marginally higher user trust but lower performance,” shows that although participants seem to report slightly higher trust in eyed robots, they performed the task faster and showed larger pupil size (a possible indicator of greater interest in the object) when they interacted with robots that did not have eyes.
The researchers conclude that humans might not need human-like machines to trust and work with them, but instead seem to collaborate better with machine-like, eyeless robots.
Pilacinski states: “We concluded that the cobot’s gaze may not be that important for manual collaboration,” noting that “objective metrics, such as task completion time and pupil responses, suggest more comfortable cooperation with non-eyed robots.”
The findings of the researchers from Ruhr University Bochum in Germany, the University of Coimbra, and the University of Trás-os-Montes and Alto Douro in Portugal align with recent research suggesting that anthropomorphism may be detrimental in collaborative robots.
