Personalized Wearable Interactions with WearSkill
Ovidiu-Andrei Schipor, Laura-Bianca Bilius, Ovidiu-Ciprian Ungurean, Alexandru-Ionut Siean, Alexandru-Tudor Andrei, Radu-Daniel Vatavu · 2022 · Proceedings of the 19th International Web for All Conference (W4A) · doi:10.1145/3493612.3520474
Summary
This extended abstract presents WearSkill, a web-based middleware application that enables people with upper-body motor impairments to control connected smart home devices through personalised wearable interactions. The system addresses a significant gap: most wearable devices assume standard motor abilities for touch and gesture input, creating barriers for users with conditions like spinal cord injury or muscular dystrophy. WearSkill acts as a bridge between wearable input devices (smartwatches, smart rings, and smartglasses) and Wi-Fi-connected appliances, allowing users to control things like smart lighting and media playback. The key innovation is personalised input recommendation — based on users self-reporting their specific motor impairments from eleven categories (including slow movements, spasms, low strength, tremor, poor coordination, and difficulty gripping), WearSkill recommends appropriate input modalities (touch, motion, or voice) and suitable wearable form factors. The project represents a two-year research effort including a systematic literature review of wearable accessibility, two user studies with people with motor impairments, and the software development itself. The system is built on Vue.js and Node.js using WebSocket communications, following the WISE framework the team developed for increasing wearable accessibility.
Key findings
The underlying research revealed that wearable accessibility is significantly under-studied — a systematic review found only four papers on smartwatch input for users with motor impairments, with disproportionate focus on hand gestures over other modalities, and a median of only 6 participants with motor impairments per study. A user study with 14 participants found that people with upper-body motor impairments took twice as long to produce stroke-gestures on wearable touchscreens compared to users without impairments, but — crucially — articulated mid-air motion gestures equally fast and with similar acceleration characteristics. This suggests motion input is a more equitable modality for wearable interaction. A second study with 21 participants with motor impairments found high preferences for smartwatches for smart home control and for rings and bracelets for payments and public interactions. When evaluated, WearSkill's personalised recommendations matched 85.3% of users' actual preferences for input modalities and wearable devices, demonstrating the viability of self-reported impairment data for input personalisation.
Relevance
WearSkill demonstrates a practical approach to making the rapidly growing wearable device market accessible to people with motor impairments. The finding that motion gestures are performed equally well by users with and without impairments challenges the dominant reliance on touch input in wearable design and points toward more inclusive default interactions. The personalisation approach — matching input modalities to specific impairment profiles rather than treating motor disability as a monolithic category — is a model that could be applied broadly across assistive technology. The WISE framework (diverse Wearable designs, new Input modalities, more user Studies, and Extending to other devices) provides a useful research roadmap. As a short demo paper, WearSkill has not yet been evaluated in real-world deployment with sustained use, and the 85.3% preference match, while promising, was tested in a controlled setting.
Tags: wearable technology · motor impairment · personalization · smart home · gesture input · assistive technology · middleware · ability-based design
Standards referenced: ISO/IEC 25010:2011