Adaptation gap hypothesis : How differences between users ’ expected and perceived agent functions affect their subjective impression

We describe an “adaptation gap” that indicates the differences between the functions of artificial agents that users expect before starting their interactions and the functions they perceive after their interactions. We investigated the effect of this adaptation gap on users’ impressions of artificial agents because any variations in impression before and after the start of an interaction determines whether the user feels that this agent is worth interacting with. The results showed that positive or negative signs of the adaptation gap and subjective impression scores of agents before the experiment significantly affected the users’ final impressions of the agents.