Population growth and changing family structures have increased social isolation among the elderly, making it a global issue. Fortunately, recent advancements in artificial intelligence(AI) have offer promising solutions. This study developed a chatbot integrated with the LINE application, utilizing social support technology to provide interactive engagement for the elderly. The study conducted a one-week human-computer interaction experiment in June 2023, collecting a total of 282 responses from eight older adult participants (mean age = 66 years, SD = 3.2), including four men and four women. The results revealed that interactions between the elderly and the chatbot can be classified into six categories: Agreement and Disagreement (AD, 38.1%), Questions and Answers (QA, 19.9%), Greetings and Descriptions (GD, 18.1%), Compliments and Affirmations (CA, 10.0%), No Response (NR, 7.5%), and Emotions and Feelings (EF, 6.4%). The results integrates six interaction categories to emerge three primary findings. First, elderly users perceive the chatbot as a conversational partner (GD and EF), initiating greetings, which indicates the chatbot's role as an interactive entity. Second, elderly users regard the chatbot as an educational resource (QA and CA). Interactions often conclude with praise and gratitude, distinguishing the chatbot from typical information retrieval tools. Finally, elderly users demonstrate independent thinking by evaluating chatbot information and cross-referencing it with other sources, illustrating the chatbot's role as an Interactive Learning Partner in a bidirectional learning process.

Authors:
Ya-Ling Wang, National Taiwan Normal University, Taiwan
Yi-Ting Chuang, National Taiwan Normal University, Taiwan

About the Presenter(s)
Yi-Ting,CHUANG is a master's student in the Department of Adult and Continuing Education at the College of Education, National Taiwan Normal University.

See this presentation on the full schedule – Monday Schedule

Bookmark