A recent study of social robotics has found that humans, in general, prefer interacting with faulty robots far more than with robots that function and behave flawlessly. This finding could impact on the future design approach to robot manufacture and programming.
The focus of the study was to explore how people react to robots that exhibit faulty behavior, and contrasting this to how people interact with near-perfectly performing robots. The results surprised the researchers and showed that the human participants took a far stronger liking to a faulty robot than to a robot that appears to interact flawlessly. The study was undertaken at the enter for Human-Computer Interaction, University of Salzburg, Austria. The robots used were so-termed ‘social robots’; a social robot is an autonomous robot that interacts and communicates with humans or other autonomous physical agents by following social behaviors and rules attached to its role.
READ MORE: Robot uses deep learning to write its own music
The robots used in the study (‘perfect’ and ‘flawed’) were tested out with a range of different people. The researchers measured the robot’s likability, anthropomorphism and perceived intelligence. The human subjects also competed questionnaires.
Commenting on the study outcome, lead researcher Dr. Nicole Mirnig told Phys.org: “Our results show that decoding a human’s social signals can help the robot understand that there is an error and subsequently react accordingly.” She went onto say: “Our results showed that the participants liked the faulty robot significantly more than the flawless one. This finding confirms the Pratfall Effect, which states that people’s attractiveness increases when they make a mistake.”
READ MORE: Robotic advances for MedTech
In social psychology, the Pratfall Effect is the tendency for attractiveness to increase or decrease after an individual makes a mistake, depending on the individual’s perceived ability to perform well in a general sense. It seems, therefore, that our fondness for the little characteristics and errors in our fellow-humans is something we’d like to see replicate with future robots.
Social robotics is a relatively new field, but it’s an important one as robots become commonplace. This will lead to different reactions from people, and trying to second-guess these reactions will need to become part the interactional quality of a robot’s core programming. The field also includes searching for insights into human behavior, from the perspective of the machine and observing how people engage with a robot.
The research has been published in the journal Frontiers in Robotics and AI, 2017, in a paper titled “To Err Is Robot: How Humans Assess and Act toward an Erroneous Social Robot.”
