Faces, voices, mimicry, empathy and unpredictability.

The acceptance of robots in the workplace is not a problem of efficiency, but rather one of sociology. Researchers from Northwestern University’s Kellogg School of Management and the Harvard Business School have recently performed a review of which design characteristics affect people’s feelings about robots, and have found a range of issues related to the elicitation of human emotions.

Design characteristics which were found to help achieve emotional acceptability included a human-like face, particularly a "baby face". Child faced robots were considered more expressive and less likely to threaten. Another important design characteristic was voice. People preferred airline-reservation robots with humanlike speech patterns rather than synthetic speech patterns, and felt emotionally closer to robots whose voices matched their own gender. Gestures were also found to be important. People preferred robots that nodded when they nodded, and blinked when they blinked. Some gestures such as the robot touching its face or folding its arms lead to mistrust, while mimicking gestures instead facilitated rapport.

Finally, the most counterintuitive way to enhance robot acceptance was found to be to make them slightly unpredictable. Real people have good and bad days, thus robots should as well. In a five-month study of toddlers’ responses to a robotic child-care assistant, the children interacted most positively with the robot when it behaved with some variability. When the robot behaved predictably, the interactions deteriorated. Unpredictable people capture our attention, and it seems that unpredictable robots do the same.

The robots of the future are therefore likely to have faces, voices, mimicry, empathy and unpredictability…

Should be fun…