Grasping the complexity of UX in the Long-term in Interaction with Embodied Conversational Agents

To comprehensively evaluate the experience of interaction between humans and ECAs (Embodied Conversational Agents), it is essential to consider a complex entanglement of dimensions that dynamically develop over time. This paper provides a review of the available methods to evaluate the human-ECAs interaction, and discusses the relation between this experience of use over time. The research was carried out through a systematic literature review, on 5 databases: Scopus, ACM,Web of Science, PubMed and IEEE Xplore. Findings reveal that over 147 publications on human-ECAS interaction, only 13 address the experience evaluation. Of these, only 3 face evaluate the user experience with ECAs over time. The majority of these evaluation methods concentrate on momentary experiences, neglecting the long-term perspective. The paper provides a map of methods to evaluate several user experience dimensions of interactions with ECAs, highlighting the importance to apply these methods not only on the momentary interaction but repeatedly over time.