Austrian and British engineers appreciated how much people enjoy communicating with robots that work perfectly, and those whose programs are flawed. Read more about the research in the journal Frontiers in Robotics and AI.
“The behaviors that emerge in unexpected situations during the experiment are often overlooked or simply ignored. The reason for this is in the very nature of scientific research, subject to strict order. However, it seems to us that errors in communication between robots and humans can help us improve the quality of their interaction,”commented Nicole Miernig, one of the authors of the work, a researcher from the University of Salzburg.
To study how "errors" in the robot's behavior affect its communication with humans, scientists intentionally added them to the robot's program. The participants in the experiment were asked to rate the attractiveness of the robot, its anthropomorphism (the degree of similarity with a person) and intelligence. The scientists also observed the reaction of the subjects to the robot's errors. The results showed that robots with program disabilities were not considered much less intelligent or human-like, but were rated much higher on the attractiveness scale.
The authors of the study believe that the creators of robots should include in them the ability to make mistakes and learn from them. “Studying the reasons for the imperfection of robots will allow us to create robots with more realistic characters and a more natural style of communication with humans,” added Myrnig.
