Building rapport with extraverted and introverted agents

Psychology research reports that people tend to seek companionship with those who have a similar level of extraversion, and markers in dialogue show the speaker’s extraversion. Work in human-computer interaction seeks to understand creating and maintaining rapport between humans and ECAs. This study examines if humans will report greater rapport when interacting with an agent with an extraversion/introversion profile similar to their own. ECAs representing an extrovert and an introvert were created by manipulating three dialogue features. Using an informal, task-oriented setting, participants interacted with one of the agents in an immersive environment. Results suggest that subjects did not report the greatest rapport when interacting with the agent most similar to their level of extraversion. Introduction People often seek companionship with those who have a personality similar to their own [11]. There is evidence that personality types are borne out in dialogue choices [1]. Humans are uniquely physically capable of speech, and tend to find spoken communication as the most efficient and comfortable way to interact, including with technology [2]. They respond to computer personalities in the same way as they would to human personalities [10]. Recent research has sought to understand the nature of creating and maintaining rapport—a sense of emotional connection—when communicating with embodied conversational agents (ECAs) [2, 8, 14]. Successful ECAs could serve in a number of useful applications, from education to care giving. As human relationships are fundamentally social and emotional, these qualities must be incorporated into ECAs if human-agent relationships are to feel natural to users. Research has been focused on the development and maintenance of rapport felt by humans when interacting with an ECA [11] and in developing ECA personalities [3]. However, questions remain as to which agent personality is the best match for developing rapport in human-ECA interactions. In this study, two agents representing an extravert and an introvert were created by manipulating three dialogue features. Using a task-oriented but informal set-