Lose the tension with Hyperbot: an applied communications research on the effects of chatbot empathy and multiple cues in communicating for hypertension screening

Hypertension is the single most important risk factor for cardiovascular diseases (CVD). It is called a “silent killer” due to the lack of symptoms, and there are generally low levels of hypertension awareness observed among the population. Chatbots have demonstrated potential to improve self-manage...

Full description

Bibliographic Details
Main Authors: Song, Janelle Yu Wei, Lim, Cheryl Yu Yee, Tien, Deborah Nien Ci, Tan, Derrick Rui Yang
Other Authors: Kay (Hye Kyung) Kim
Format: Final Year Project (FYP)
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/174414
Description
Summary:Hypertension is the single most important risk factor for cardiovascular diseases (CVD). It is called a “silent killer” due to the lack of symptoms, and there are generally low levels of hypertension awareness observed among the population. Chatbots have demonstrated potential to improve self-management behaviours for diseases, but there have been limited studies conducted for hypertension screening advocacy. Our study aims to launch a pilot campaign that examines the influence of empathetic language and multiple cues in a chatbot on young adults’ intentions to attend hypertension screening. A study was first conducted among our target audience to understand their perceptions of popular social media platforms. This survey (N = 77) revealed that Telegram Messenger was perceived to be the most media-rich and most frequently used platform. We then designed four chatbots in our pilot campaign and deployed them on Telegram to test for potential effects of empathetic language and multiple cues on encouraging intentions to engage in our desired health behaviour, alongside a control group. Results (N = 150) showed that while the chatbots effectively increased participants’ intentions, empathetic language and multiple cues did not significantly impact these intentions. Practical implications in health communication from our study were discussed.