Text Entry for Mobile Devices and Users with Severe Motor Impairments: HandiGlyph, a Primitive Shapes Based Onscreen Keyboard
Mohammed Belatar, Franck Poirier · 2008 · Proceedings of the 10th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '08) · doi:10.1145/1414471.1414510
Summary
This paper presents HandiGlyph, a text input method for mobile devices designed for users with severe motor disabilities such as Cerebral Palsy, Amyotrophic Lateral Sclerosis, Locked-In Syndrome, and quadriplegia. The system is built on the UniGlyph principle: each letter of the alphabet is decomposed into primitive shapes that visually resemble the capital Latin character (e.g., the primitives for a letter are chosen based on analogy with its shape). This means the keyboard has only three primitive keys plus a command key — a total of four keys — dramatically reducing the motor demands compared to full QWERTY layouts. The interface combines these primitive keys with two display areas: a disambiguation list (words matching the exact primitive sequence entered) and a completion list (words starting with the entered sequence), both drawn from a 40,000-word French dictionary. Scanning is used for navigation — the focus automatically cycles through the keys and display areas, and the user performs simple clicks (short or long) on a two-state sensor such as a push-button, breath sensor, or head movement switch to make selections.
Key findings
An expert-level evaluation was conducted with a 26-year-old man with Locked-In Syndrome at the Kerpape Rehabilitation Center in France, over five sessions copying randomly selected French sentences (30-40 characters each). The average KSPC (Keystrokes Per Character) across all sessions was 1.5, indicating reasonable efficiency given the error correction overhead. The MSD (Minimum String Distance) error rate dropped from 4.58% in session 1 to 0% in sessions 3 and 4, showing rapid learning. Average input rate was 2.37 words per minute with a maximum of 3.18 WPM. Scanning delay evolved dramatically from 754.8ms in session 1 to 118ms by session 5, demonstrating the adaptive scanning delay algorithm's effectiveness in matching the user's improving speed. The user spent more time on the disambiguation list than the completion list, preferring to enter complete primitive sequences before selecting words. The participant expressed strong satisfaction, stating "wow, I want this system" despite daily use of other AAC systems, and indicated interest in combining HandiGlyph with text-to-speech.
Relevance
HandiGlyph addresses a critical gap in accessible text entry: most mobile text input methods are designed for able-bodied users and remain unusable for people with severe motor impairments. The system's design reflects key accessibility principles — using analogy with familiar letter shapes to minimize learning burden, requiring only a two-state sensor input that accommodates the widest range of motor abilities, and providing adaptive scanning delays that automatically calibrate to the user's capabilities. For AAC practitioners, the approach of decomposing characters into a small set of primitive shapes offers a middle ground between full scanning keyboards (slow but simple) and ambiguous multi-tap methods (faster but cognitively demanding). The successful evaluation with a person with Locked-In Syndrome — one of the most severe motor conditions — demonstrates the system's potential for users across the motor disability spectrum.
Tags: text entry · motor disability · on-screen keyboard · AAC · scanning input · locked-in syndrome · mobile accessibility · assistive technology