The Smiling Chatbot: Investigating Emotional Contagion in Human-to-Chatbot Service Interactions

Significant advances in the field of artificial intelligence have given a strong tailwind to the spread of technologies based on it in recent years. AI technologies are increasingly influencing private contexts but also the service sector where more and more service encounters are handled by chatbots. In part, however, the spread of chatbots and the associated focus on their functional advantages seem to contradict extant research findings that service interactions are also clearly driven by emotional components. A central role is played here, for example, by so-called emotional contagion (i.e., the unconscious transfer of emotions from employees to customers). Against the background that this contagion with usually positive emotions can lead to a better evaluation of the service transaction in the further course, it was the goal of the present thesis to investigate the effects of positive displayed emotions of a chatbot. For this purpose, six consecutive studies were conducted. The results clearly show that the expression of positive emotions by a chatbot enriches the service interactions by transmitting positive emotions that, in the further course, do also lead to a better evaluation of the service experienced. Moreover, it is shown that these emotional reactions are dependent on the customer’s personality and the chatbot’s appearance, expressed through an avatar. Furthermore, the results suggest that the reactions occur also during critical service recovery interactions.