Communication and Personality: how COVID-19 government chatbots express themselves

Abstract

With the COVID-19 pandemic, governments have used applications to help citizens to be informed about the disease. Some applications use chatbots, which despite the advances, still represent challenges for researchers and designers in the domain of human-computer interaction (HCI). Therefore, this paper aims to evaluate three Brazilian government chatbots focused on COVID-19, considering two perspectives: communication (which involves communicability and language) and personality. For this, we used the Semiotic Inspection Method (SIM) to evaluate the chatbots' communicability and best practice guidelines for evaluating the chatbots' language and personality. As a result, we show that chatbots make little use of communicability strategies. In addition, they do not convey personality traits. We also discussed the relationship between personality and the designer's self-expression, making room for potential future work in the area. The research contributes to the extent that the study of COVID-19 chatbots in the current pandemic scenario is important and presents a methodology for assessing the language and personality of chatbots. © 2021 ACM.