Dialoging Resonance: How Users Perceive, Reciprocate and React to Chatbot’s Self-Disclosure in Conversational Recommendations

Using chatbots to deliver recommendations is increasingly popular. The design of recommendation chatbots has primarily been taking an information-centric approach by focusing on the recommended content per se. Limited attention is on how social connection and relational strategies, such as self-disclosure from a chatbot, may influence users’ perception and acceptance of the recommendation. In this work, we designed, implemented, and evaluated a social chatbot capable of performing three different levels of self-disclosure: factual information (low), cognitive opinions (medium), and emotions (high). In the evaluation, we recruited 372 participants to converse with the chatbot on two topics: movies and COVID-19 experiences. In each topic, the chatbot performed small talks and made recommendations relevant to the topic. Participants were randomly assigned to four experimental conditions where the chatbot used factual, cognitive, emotional, and adaptive strategies to perform self-disclosures. By training a text classifier to identify users’ level of self-disclosure in real-time, the adaptive chatbot can dynamically match its self-disclosure to the level of disclosure exhibited by the users.

Introduction. As chatbots are becoming increasingly popular across various digital platforms, chatbot-human conversations have evolved into more complex and extended exchanges that often invoke disclosing personal information and emotional experiences. In recent years, developments in natural language processing have empowered chatbots with a greater capacity to engage in social and emotional conversations. In addition, endeavors have been made to create and test persuasive or recommendation chatbots that initiate recommendations to users to adopt. A study showed that people However, these studies so far have only emphasized the recommended content itself by providing more information and explanation about it. This information-centric approach neglects and underutilizes the chatbot’s characteristics relating to social identity and human-chatbot relationship. As many chatbots are personified and equipped with social conversation capabilities, there have been few investigations concerning how social connection and relational strategies would influence persuasion and recommendation effectiveness.

Discussion / Conclusion. This study experimentally tested how people reacted to chatbot’s different levels of self-disclosure with a text-based sociable recommendation chatbot. We showed that peoples’ self-disclosure levels were positively correlated to the chatbot’s self-disclosure levels. Also, we found that higher levels of self-disclosure led to more engaging conversations and warmer bot perception. Lastly, emotional self-disclosure significantly enhanced people’s enjoyment and attitude to a chatbot’s recommendations. We believe our work provides a better understanding of how a bot’s self-disclosure can be leveraged to encourage users’ self-disclosure, improve users’ perception of a chatbot, and enhance the persuasiveness of the recommendation.