We performed an empirical study exploring people's interactions with an embodied conversational agent (ECA) while performing two tasks. Conditions varied with respect to 1) whether participants were allowed to choose an agent and its characteristics and 2) the putative quality or appropriateness of the agent for the tasks. For both tasks, selection combined with the illusion of further customization significantly improved participants' overall subjective impressions of the ECAs while putative quality had little or no effect. Additionally, performance data revealed that the ECA's motivation and persuasion effects were significantly enhanced when participants chose agents to use. We found that user expectations about and perceptions of the interaction between themselves and an ECA depended very much on the individual's preconceived notions and preferences of various ECA characteristics and might deviate greatly from the models that ECA designers intend to portray. Au...
Jun Xiao, John T. Stasko, Richard Catrambone