A Proposal to Extend the Modeling Language for Interaction as Conversation for the Design of Conversational Agents

Conversational agents have become a topic of growing interest in recent years. Their increasing popularity offers opportunities and challenges for Human-Computer Interaction (HCI). Among them, there is a need for more research into whether existing HCI dialogue models apply to conversational agents. Our research focuses on MoLIC (Modeling Language for Interaction as Conversation), a design phase dialogue model based on Semiotic Engineering theory, which allows designers to represent interaction as conversations between a system and its users. Previous studies have pointed out MoLIC’s limitations in modeling conversational agents. In this article, our goal is to propose and evaluate an extension to MoLIC to broaden its expressiveness and allow it to be applied to the context of conversational agents. We describe the new elements and adaptations proposed and illustrate how they can be used to model relevant aspects of conversational agents. To evaluate the extension proposed we conducted a case study, in which we applied the extended version of MoLIC in a reverse engineering modeling of an existing chatbot, the chatbot for the Superior Electoral Court (TSE) of Brazil. Our results show that the proposed extension was useful and necessary for describing the TSE’s interaction model. This work brings research contributions to the field of HCI, and in particular to the research and development of conversational agents, as well as research on MoLIC and Semiotic Engineering.