Study reveals wheelchair users' phone-carrying patterns beyond pockets

A new study from Yunzhi Li and Patrick Carrington, accepted at MobileHCI 2026, systematically explores how wheelchair users carry their smartphones — a crucial but overlooked factor for mobile context-awareness. Using a mixed-methods approach (survey of 91 participants and in-depth interviews with 15), the researchers discovered that wheelchair users rely on a wide range of carrying locations far beyond typical pants pockets. Common placements include wheelchair-mounted phone holders (e.g., on the frame, under the seat), around-body pouches, and on-wheelchair straps, each chosen based on the user's physical abilities, wheelchair design, and the immediate social or activity context. For example, some users prefer front-mounted holders for easy access while navigating, while others use back-mounted pouches for stability during transfers. The study also found that factors like device size, activity (e.g., using voice commands vs. typing), and social settings (e.g., formal vs. casual) influence where and how the phone is carried.

The findings have direct implications for building inclusive mobile sensing systems. Existing context-aware algorithms — which infer user activity from phone sensors — are typically trained on data from ambulatory users carrying phones in pockets or hands, leading to poor accuracy for wheelchair users. By recognizing that carrying location itself can act as a proxy for user context (e.g., a phone mounted on the wheelchair armrest may indicate active driving vs. a phone on the lap during transfers), researchers can design more robust and personalized sensing pipelines. The paper also discusses design guidelines for app developers: for instance, allowing users to specify their phone location in setup, or using on-device ML to adapt inference models dynamically. This work not only fills a gap in HCI research but also advocates for equitable mobile experiences for people with disabilities.