← All reviews

"This really lets us see the entire world:" Designing a conversational telepresence robot for homebound older adults

Yaxin Hu, Laura Stegner, Yasmine Kotturi, Caroline Zhang, Yi-Hao Peng, Faria Huq, Yuhang Zhao, Jeffrey P. Bigham, Bilge Mutlu · 2024 · Proceedings of the 2024 ACM Designing Interactive Systems Conference (DIS '24) · doi:10.1145/3643834.3660710

Summary

This paper investigates how conversational telepresence robots can help homebound older adults interact with the external world. Homebound status — defined as rarely or never leaving home due to illness, injury, or environmental constraints — affects a growing proportion of older adults and is associated with higher mortality, functional decline, and depression. While prior research has explored telepresence robots for communication and medical tasks, little work has examined how these robots might enable exploration and meaningful experiences in remote physical environments. The researchers conducted two studies using a research through design (RtD) approach. The first was a needfinding study with eight homebound older adults (ages 78–94) living in a senior care facility. Using scenario-based video vignettes depicting a robot visiting a farmer's market, botanical garden, and lakefront, the researchers identified three desired experience categories: reminiscent experiences (revisiting familiar places and recalling past memories), exploratory experiences (learning about new environments and objects), and social participation (both active conversations with bystanders and passive observation of community life). The second study was a technology probe with 11 participants (ages 80–98) who used a prototype conversational telepresence robot — built from a Turtle Bot mobile base with a camera phone and controlled via Wizard of Oz — to remotely visit a lakefront park and botanical garden. The robot's conversational interface provided guided narrations, supported social chat, and facilitated bystander interactions through a video conferencing overlay.

Key findings

All participants preferred the robot's guidance over self-directed control, primarily because they lacked knowledge of the remote environment and confidence in operating the robot. The robot's narrative descriptions were valued for being informative and specific, though some participants wanted more professional or detailed narrations. Participants spontaneously disclosed personal histories triggered by familiar scenes — gardens prompted stories about family, lakes evoked memories of past trips with spouses, and campus settings recalled college experiences from decades earlier. Five of 11 participants gave positive overall feedback, describing the experience as wonderful, fun, and fascinating. However, five participants disliked the experience, with four refusing to continue after the first session. Challenges included lack of environmental awareness, confusion about the robot's directional controls (participants used the screen as their frame of reference rather than the robot's), difficulty comprehending the experience, and discomfort with verbal interaction. Bystander interactions revealed social boundary violations — some participants asked personal questions or made inappropriate comments to strangers, highlighting the need for social norm safeguards.

Relevance

This research opens an important design space for using conversational robots to reduce social isolation among homebound older adults — a population that is growing rapidly and faces significant barriers to community participation. The findings are particularly relevant for accessibility practitioners designing assistive technologies that go beyond functional task completion to support meaningful life experiences. The study's emphasis on the robot's proactive role — providing guidance, curating personalized experiences, and mediating social interactions — offers a model for designing AI-mediated accessibility tools. The mixed reception (nearly half of participants disliked the experience) underscores the importance of iterative design, adequate onboarding, and accommodating diverse technology comfort levels among older adults. The communication challenges observed, including hearing difficulties and turn-taking breakdowns during bystander interactions, point to critical accessibility considerations for speech-based interfaces used by older adults.

Tags: telepresence robots · older adults · homebound · social isolation · participatory design · conversational interface · assistive technology · remote exploration · aging