Help Me Please

Abstract

Robots that can leverage help from people could accomplish much more than robots that cannot. We present the results of two experiments that examine how robots can more effectively request help from people. Study 1 is a video prototype experiment (N=354), investigating the effectiveness of four linguistic politeness strategies as well as the effects of social status (equal, low), size of request (large, small), and robot familiarity (high, low) on people's willingness to help a robot. The results of this study largely support Politeness Theory and the Computers as Social Actors paradigm. Study 2 is a physical human-robot interaction experiment (N=48), examining the impact of source orientation (autonomous, single operator, multiple operators) on people's behavioral willingness to help the robot. People were nearly 50% faster to help the robot if they perceived it to be autonomous rather than being teleoperated. Implications for research design, theory, and methods are discussed.