Why Is Gesture Typing Promising for Older Adults? Comparing Gesture and Tap Typing Behavior of Older with Young Adults
Yu-Hao Lin, Suwen Zhu, Yu-Jung Ko, Wenzhe Cui, Xiaojun Bi · 2018 · Proceedings of the 20th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2018) · doi:10.1145/3234695.3236350
Summary
This study from Stony Brook University investigates whether gesture typing (gliding a finger continuously across keys to spell words) is a viable alternative to traditional tap typing for older adults on smartphones. Text entry is one of the most significant barriers older adults face with mobile devices — age-related declines in motor control, visual acuity, and cognitive processing make it difficult to accurately hit small keys on a touchscreen keyboard. The researchers conducted a controlled experiment with 28 participants: 14 older adults (ages 65-81, recruited from Senior Planet Center in New York City) and 14 young adults (ages 24-34). Critically, none of the older adults had any prior gesture typing experience, and 13 of 14 had never even heard of it. Participants completed phrase transcription tasks using both Google Keyboard gesture typing and tap typing on a Nexus 5X smartphone, with all touch events logged at the stroke level. The study used a 2x2 mixed factorial design (age group x keyboard type) with 4 blocks of 5 phrases each per condition, and employed established modeling tools including the CLC (curves, lines, corners) model to analyze gesture production time and the SHARK2 algorithm metrics for gesture shape and location accuracy.
Key findings
The headline finding is striking: gesture typing was 15.28% faster than tap typing for older adults (26.02 WPM vs 22.57 WPM), with 27.1% lower word error rate (6.79% vs 9.32%), despite participants having zero prior experience. Older adults learned gesture typing quickly, reaching over 25 WPM after just 10 phrases. The stroke-level analysis revealed why gesture typing works better for this population. In tap typing, older adults adopted a counterproductive letter-level correction strategy — pressing backspace to fix errors immediately as they appeared (41% of backspace usage was intermediate, mid-word corrections). This prevented them from benefiting from the statistical decoder's auto-correction, which operates at the word level. In gesture typing, the continuous nature of the input forced a word-level strategy — users could only check results after lifting their finger — which aligned with how the decoder works. Gesture accuracy was surprisingly preserved with aging: older adults' shape error (0.067) and location error (53.7 pixels) were comparable to young adults' (0.073 and 53.7 pixels), suggesting that degraded motor control has minimal impact on gesture trace quality. However, older adults were significantly slower at producing long gestures, with the speed gap growing dramatically with gesture length. Touch point distributions in tap typing spread much wider for older adults, confirming the well-known difficulty of precise key targeting.
Relevance
This research has direct practical implications for mobile accessibility. The finding that gesture typing outperforms tap typing for older adults — even with no prior experience — suggests that smartphone keyboards should more actively promote and teach gesture typing to older users. Currently, gesture typing is often a hidden feature that users must discover on their own. For keyboard designers, the study identifies specific improvements: allowing hybrid gesture-and-tap input for long words (since long gestures are disproportionately slow for older adults), adapting the spatial model of the decoder to account for the wider touch point distributions of older users, and designing UI cues that encourage users to trust auto-correction rather than manually correcting every letter. The finding about letter-level correction undermining tap typing performance also applies broadly — many users of all ages likely exhibit the same behavior, and keyboard design that makes auto-correction more visible and trustworthy could benefit everyone.
Tags: older adults · text entry · gesture typing · touchscreen accessibility · mobile accessibility · motor control · aging · input methods