Trust, Comfort, and Relatability: Understanding Black Older Adults’ Perceptions of Chatbot Design for Health Information Seeking

Title of the Paper

Trust, Comfort and Relatability: Understanding Black Older Adults’ Perceptions of Chatbot Design for Health Information Seeking

Paper Information

• Subject Area: User experience and design optimization in artificial intelligence and health information technology
• Keywords: Chatbot, Health Information, Older Adults, Race, Identity, Interactive Virtual Assistant, Diary Study, Trust, Relatability

Research Background and Issues

• What problems or challenges did the authors identify?

  • Due to differences in factors such as race and age, Black older adults face challenges in accepting health information tools like chatbots.
  • This user group has lower trust and comfort levels with online health information platforms, which affects their ability to adopt and use chatbots for health information.
  • There is a lack of in-depth research on how the design of chatbots reflecting race, age, and gender characteristics impacts the experience of Black older adults.

• Why is this issue important?

  • Chatbots have evolved into cost-effective, real-time health information resources that can help technologically inexperienced user groups (such as Black older adults) overcome barriers to accessing information.
  • Designing health information tools for marginalized groups requires consideration of cultural characteristics, technological accessibility, and user experience to promote equity in technology adoption.

• Research Motivation and Related Work

  • To address the research gap in chatbot design that considers racial and age differences, the authors conducted an empirical analysis of Black older adults’ perceptions, exploring ways to enhance trust, comfort, and relatability.
  • Building on prior research on chatbot anthropomorphism and trust factors, the study further integrates chatbot appearance design with the health information needs of the Black community.

Solution

• What methods or solutions did the authors propose?

  • Conducted a diary and interview study, collecting interaction experiences and design feedback from 30 Black older adults living in low-income communities.
  • Analyzed how the design of chatbot personas (race, age, and professional identity) influenced participants’ trust, comfort, and acceptance of chatbots.

• What is innovative about this solution?

  • Proposed chatbot design recommendations that fully consider racial and age characteristics, focusing on dimensions such as credibility, relatability, and cultural relevance.
  • Beyond racial characteristics, the study explored how other design elements, such as attire, language style, and professional identity, enhance user trust and comfort.

• What are the implementation steps and key technologies used?

  • Recruited participants and introduced them to chatbot products for hands-on experience.
  • Used physical paper diaries to document participants’ daily health information needs and their feelings during interactions with chatbots.
  • Conducted follow-up interviews to gather feedback on chatbot personas with different identities and explored factors influencing user trust, relatability, and comfort.

Research Findings

• What specific findings were obtained?

  • The study found that appropriate persona design (including race, age, gender, and professional image) effectively increased the acceptance of chatbots among Black older adults.
  • Racial congruence provided emotional relatability for users but did not directly eliminate their historical distrust of medical technology.
  • User trust in chatbots was primarily driven by information transparency and professional identity rather than simple racial matching.

• What advantages does it have compared to existing solutions?

  • Offers more detailed design guidelines, emphasizing multi-layered identity considerations such as race, gender, and age.
  • Addresses the historical distrust of medical technology among Black older adults by proposing recommendations that include design transparency and cultural relevance.

• What were the experimental or evaluation results?

  • Users were more willing to interact with chatbots that wore professional attire and represented credible professional identities (e.g., “doctor” or “nurse” personas).
  • Users expressed a preference for seeking face-to-face professional consultations for complex or high-risk health issues rather than relying on chatbots.
  • Relying solely on a “Black appearance” to build trust in chatbots might be perceived by some users as pandering or deceptive.

• Limitations and Future Directions

  • Limitations:
    • The majority of participants were female, and the gender imbalance may have influenced the results.
    • Some participants were familiar with technology, leaving the behavior of a broader group of low-tech users unclear.
  • Future Research Directions:
    • Further studies on user groups with diverse genders and non-binary identities.
    • Design more transparent health chatbot platforms that address more complex and high-risk health issues.
    • Conduct larger-scale quantitative studies to validate the effects of racial characteristics in design on building user trust.