Human Considerations In The Application Of Cognitive Decision Models For Hri

In order for autonomous robots to succeed as useful teammates for humans, it is necessary to examine the lens through which human users view, understand, and predict robotic behavior and abilities. To further study this, we conducted an experiment in which participants viewed video segments of a robot in a task-oriented environment, and were asked to explain what the robot was doing, and would likely do next. Results showed that participants\u27 perceived knowledge of the robot increased with additional exposures over time; however participant responses to open-ended questions about the robot\u27s behavior and functions remained divergent over multiple scenarios. A discussion of the implications of apparent differences in human interpretation and prediction of robotic behavior and functionality is presented. © 2013 Springer-Verlag Berlin Heidelberg