Despite the rise of artificial intelligence (AI) its practical applications, including in health promotion, AI still remains relatively new to older adults. This study explores gender-based differences in the conversations of older individuals discussing health with AI chatbots. We recruited 31 Taiwanese older adults (52% female), averaging 67.74 years (SD = 5.52, range 59-79). All participants were required to use the LINE messaging application to chat with a chatbot developed by the researchers about health-related topics throughout the study period of five days. The study resulted in a total of 364 chat records from older adults (an average of 15.79 records/person). After the researchers encoded these records, they found that the data could be categorized into nine groups: diet (36%), exercise and physical activity (10%), diseases and symptoms (9%), body image or appearance (3%), mental health (9%), lifestyle habits (12%), greetings (4%), agreeing/affirmation with the AI Chatbot (13%), and off-topic (non-health-related conversations (5%). Furthermore, through a chi-square analysis of the encoding results, it revealed that messages related to diet predominantly came from female older adults, whereas topics related to lifestyle habits were more commonly contributed by male older adults. From the summarized results, it is evident that older adults are more interested, curious, or willing to share their thoughts on health issues related to diet. This study not only reveals new research applications for AI chatbots but also extends the use of new technology to older adults, providing valuable insights for future researchers.

Authors:
Ya-Ling Wang, National Taiwan Normal University, Taiwan
Chih-Chi Liu, National Taiwan Normal University, Taiwan

About the Presenter(s)
Mr Chih-Chi Liu is a University Doctoral Student at National Taiwan Normal University in Taiwan

See this presentation on the full schedule – Tuesday Schedule

Bookmark