As an increasing number of robots have been designed to interact with people on a regular basis, research into human-robot interaction has become more widespread. At the same time, little work has been done on the problem of longterm human-robot interaction, in which a human uses a robot for a period of weeks or months. As people spend more time with a robot, it is expected that how they make sense of the robot - their “cognitive model” of it - may change over time. In order to identify factors that will be critical to the future development of a quantitative cognitive model of long-term human