Exploring Older Adults Perceptions of Personalities of LLM-Powered Conversational Companions

The health impacts of social isolation and loneliness are well-documented. Research has linked them to an increased risk of depression, anxiety, cardiovascular diseases, and dementia, among other conditions. Older adults are particularly vulnerable to social isolation and loneliness, often due to factors such as reduced mobility with age. Recent studies have shown promise in using conversational agents (CAs) to reduce loneliness among older adults. Advances in large language models (LLMs) have enhanced these systems by enabling more natural, human-like interactions. However, little is known about how CA personalities influence user experiences, despite personality being a key factor in human conversation and socialization. To explore this, we developed a smart speaker-based CA powered by an LLM and conducted a two-phase user study consisting of in-lab sessions and home deployments. We explored older adults' perceptions of different CA personalities and their impact on interaction experiences through semi-structured interviews. Participants shared their preferences, rated the agent's personalities, and maintained diaries to document their experiences. Our findings show that participants could distinguish between different personality characteristics and had varying preferences for different personalities during both short-term and long-term interactions. We offer insights into how different aspects of certain personality traits of a conversational agent can affect the user experience and highlight the importance of enabling users to personalize their companion conversational agents to enrich their experience of socializing with such agents.