We are delighted to announce that CUI 2023 features two keynotes from distinguished academics: Effie Law and Catherine Pelachaud.
Effie Law: Trust and Emotion Resilience for Customer Service Chatbots with Breakdowns
Trust, among others, is a decisive factor for determining users’ acceptance and adoption of chatbots. Trust is even more critical when essential services such as finance and healthcare are delivered through such AI-infused digital agents. Our recent research work has investigated how different factors influencing trust in a banking chatbot. They include the chatbot’s humanlikeness and conversational performance (i.e., breakdown with/without repair), the user’s attributes, the nature of the task performed with the chatbot, and the timing of the breakdown. Furthermore, we have examined how emotional responses are related to trust in the chatbot. A series of crowdsourcing studies have been conducted with hundreds of users. Results suggest that trust and emotion resilience prevail. Users tend to forgive the imperfections of the chatbot when it behaves normally again after a breakdown. These intriguing findings can have significant implications for the design of customer service chatbots.Prof. Effie Lai-Chong Law is full professor of Human-Computer Interaction (HCI) at the Department of Computer Science, Durham University. She obtained her PhD in psychology from the University of Munich (LMU), Germany and became a research fellow at ETH Zurich, Switzerland. Before joining Durham, she worked at School of Informatics, University of Leicester, UK. Effie’s main research focus is HCI methodologies. Her recent research foci are conversational AI, mixed reality, and multisensory emotion recognition. Effie was the chair of two EU COST Actions on HCI design and evaluation methodologies: MAUSE and TwinTide. She has played a leading role in a number of EU and national projects, including the running ARETE (Augmented Reality Interactive Educational Systems; 2019-2023), and UKRI Trustworthy Autonomous Systems (TAS) Verifiability Node (2020-2024). Effie’s main contribution to these projects was applying the Human-centred Design approaches to the design and evaluation of the bespoke technologies to ensure their use quality. Effie has authored more than 200 peer-reviewed papers. Apart from her associate editor role in several reputed HCI journals, she has served as a scientific chair of the Usability and UX committee for CHI’22 and CHI’23.
Catherine Pelachaud: Adaption mechanisms for Socially Interactive Agents
During an interaction, humans continuously adapt their behaviors at different levels involving formality of language, imitation, synchronization, etc. We are working on modeling several of such adaptation mechanisms into Socially Interactive Agents, i.e. virtual agents capable of interacting socially with human partners. We have developed models that act at the conversational strategies level or at the multimodal signals one. These models drive the behaviors of these agents. They were evaluated through perceptive studies where human participants interact with them in real-time. In this talk I will present these different adaption mechanisms, the architecture of the human-agent platform in which they are implemented and the evaluation studies we conducted.Catherine Pelachaud (CNRS-ISIR) is Director of Research in the laboratory ISIR, Sorbonne University. Her research interest includes socially interactive agent, nonverbal communication (face, gaze, gesture and touch), and adaptive mechanisms in interaction. With her research team, she has been developing an interactive virtual agent platform, Greta, that can display emotional and communicative behaviors. She has participated in the organization of international conferences such as IVA, ACII and AAMAS. She is and was associate editors of several journals among which IEEE Transactions on Affective Computing, ACM Transactions on Interactive Intelligent Systems and International Journal of Human-Computer Studies. She is co-editor of the ACM handbook on socially interactive agents (2021-22).