Slipping and drifting: using older users to uncover pen-based target acquisition difficulties
Karyn A. Moffatt, Joanna McGrenere · 2007 · Proceedings of the 9th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '07) · doi:10.1145/1296843.1296848
Summary
Moffatt and McGrenere present an empirical study at the University of British Columbia investigating the underlying causes of pen-based target acquisition difficulty on a Tablet PC, with explicit attention to how age affects targeting ability. Thirty-six right-handed participants — twelve young (18–54), twelve "pre-old" (55–69), and twelve old (70–85) — completed two tasks chosen to expose different layers of difficulty: a multi-dimensional Fitts' tapping task with three target widths, three amplitudes, and eight angles; and a menu-selection task using three twelve-item menus. Standardized motor-skill instruments (Digit Symbol Substitution, Purdue Pegboard, and a 9-hole steadiness tester) were administered alongside the North American Adult Reading Test to confirm baseline ability and English fluency. The study addresses three limitations the authors identify in prior pen-input research: a narrow focus on young, healthy participants; reliance on a single highly constrained task; and a tendency to compare novel techniques against the status quo without first understanding why basic point-and-tap fails. The classification of slips, near misses, not-so-near misses, accidental taps, and other errors is adapted from Trewin, Keates, and Moffatt's earlier mouse work to fit pen interaction, and a parallel slip/correct-menu-miss/incorrect-menu-miss scheme is used for the menu task.
Key findings
Three primary sources of pen target-acquisition difficulty emerged. First, slipping — the pen lands on the target but slides off before lift — increased significantly with age and was much more severe on small targets, while younger users almost never slipped; slips were short, averaging only 10–12 pixels (2.4–2.9 mm). Second, drifting between adjacent menus, caused by the Tablet PC switching menus when the pen hovers above the screen, affected 35 of 36 participants and significantly slowed trial times; older users drifted disproportionately, and 31 participants compounded the problem by re-tapping the desired menu, which actually closed it. Drifting did not improve with practice. Third, in correct-menu misses, 82% landed on the item directly below the target and 62% on the top two pixels of that item — a tap on this top sliver was eleven times more likely to be intended for the item above. The authors propose design fixes: combining cursor freezing with a Bubble-Cursor-style area cursor to absorb slips; disabling or delaying hover-based menu switching, and disabling re-tap-to-close, to prevent drifting; and either shifting the motor target of each menu item up by two pixels or deactivating that top sliver entirely. A central methodological finding is that observing older users surfaced drifting and missing-just-below — problems that turned out to affect all ages but would have been missed with young-only participants.
Relevance
For practitioners working on pen, stylus, and touch input — and increasingly on any pointing interaction used by older adults or people with motor impairments — this paper is a useful primer on what can go wrong beneath the surface of "point and tap." The drifting and missing-just-below findings have direct relevance to current touchscreen menu and dropdown design, especially the still-common pattern where hovering over an adjacent menu auto-switches the open menu. The methodological contribution is at least as important as the specific findings: the paper is a clear case study for inclusive-design researchers in how recruiting older or disabled participants reveals usability problems that affect everyone but are masked by the speed and recovery of younger users. Limitations include the use of a single Tablet PC platform, the absence of any follow-up evaluation of the proposed designs, and the surprisingly small differences observed between the pre-old and old age groups, which suggests more nuanced participant stratification may be needed in future work.
Tags: target acquisition · pen-based interaction · stylus · tablet PC · older adults · aging · inclusive design · universal usability · Fitts' law · motor accessibility · input methods · menu interaction · usability testing