Towards Accessible Touch Interfaces
Tiago Guerreiro, Hugo Nicolau, Joaquim Jorge, Daniel Gonçalves · 2010 · Proceedings of the 12th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2010) · doi:10.1145/1878803.1878809
Summary
This paper presents an empirical evaluation of touch screen interaction techniques with 15 tetraplegic users, aiming to provide evidence-based guidance for designing accessible mobile touch interfaces. The authors recognized that while touch screen devices eliminate the need for physical button strength, they introduce new challenges for people with limited motor control — particularly the precision required to select on-screen targets without tactile feedback or physical stability. The study compared four interaction techniques: Tapping (selecting a target directly), Crossing (dragging across a target boundary), Exiting (gesturing from a target toward a screen edge), and Directional Gesturing (performing gestures anywhere on screen). Each technique was tested across three target sizes (7mm, 12mm, and 17mm) and multiple screen positions (corners, edges, and middle areas). The experiment used a QTEK 9000 PDA with a resistive touch screen, and participants — 13 male and 2 female, aged 28 to 64, with cervical spinal cord injuries ranging from C1 to C6-C7 — interacted stylus-free using their hands or fingers. The researchers measured task error rate, precision, movement error, and movement time across 2,340 total actions per participant. The study fills an important gap in the literature, as prior work on touch screen accessibility for motor-impaired users had been largely limited to able-bodied participants or users with induced impairments rather than people with actual motor disabilities.
Key findings
Tapping proved to be both the most preferred technique and among the most accurate, with error rates comparable to Crossing and significantly better than Exiting. The optimal target size was 12mm, which balanced accuracy with efficient use of screen space. Screen position had a strong effect: edges and corners provided physical stability that improved precision, while targets in the middle of the screen and at the top (farthest from arm support) produced higher error rates. For Directional Gesturing, diagonal gestures were significantly more error-prone than cardinal directions, and edge-supported gestures were straighter and faster. A key finding was that users tapped more accurately near their arm support point, and the first contact with the screen tended to be closest to this support. Crossing showed similar accuracy to Tapping but required more complex motor planning. Exiting performed worst overall, particularly at smaller sizes. The 12mm target at medium size offered the best trade-off, with error rates still high (above 10%) even at the largest 17mm size — indicating that touch interfaces inherently challenge motor-impaired users and require careful parameterization.
Relevance
This research provides foundational empirical data for designing inclusive touch interfaces, demonstrating that standard Tapping — the most common interaction paradigm — can work for motor-impaired users when properly parameterized. The finding that a unified interface can serve both able-bodied and motor-impaired users challenges the assumption that separate, specialized interfaces are always needed. For practitioners, the key takeaways are: use minimum 12mm touch targets, leverage screen edges and corners for critical controls, avoid placing important targets in the top-center area of mobile screens, and prefer simple tap interactions over gesture-based alternatives. While the study used a 2010-era PDA, the principles about screen positioning, target sizing, and the benefits of physical edge support remain highly relevant to modern smartphone and tablet design.
Tags: touch screen accessibility · motor impairment · tetraplegia · mobile devices · interaction techniques · target acquisition · inclusive design