Today's commodity software is ignorant of users' motor abilities. This places the burden of adaptation on people rather than on technology, and forces users either to struggle with off-the-shelf input devices or to procure specialized assistive devices that are expensive, hard to acquire, and infused with social stigmas. Recent approaches like universal design, inclusive design, and design for all acknowledge this problem, but promote an unachievable "one size fits all" ideal. Universal solutions are not the answer! Rather, the answer is the opposite, that is to say highly individualized user interfaces aware of and adapted to users' motor abilities. Such interfaces can enable people to use cheap readily-available everyday input devices like mice, touch pads, and trackballs when controlling desktop software. Creating software that matches users' abilities is the focus of the PI's concept of ability-based design, and this research extends the PI's prior work on ability-based design with new software tools for advancing this nascent methodology. These tools will focus on measuring and modeling target acquisition and text entry, two fundamentals of computer input that demand fine motor control, making them key challenges for people with motor impairments. The work will consist of multiple stages. The PI will investigate standard and custom psychomotor movement models for people with motor impairments. He will create and validate a Mouse Perturber tool that injects kinetic noise into an unimpaired movement stream to create simulated motor-impaired performance, making early-stage testing of ability-based prototypes easier. And he will create an Input Observer tool that can rigorously quantify users' performance "in the wild," which requires the inferring of intention outside directed laboratory tasks. He will then develop CAMA, a tool for motor-ability assessment that uses off-the-shelf input devices, and TASK, a tool for quickly creating task models from real user interfaces. Finally, he will integrate CAMA and TASK into a package for assistive technology providers to perform assessment, prediction, and the creation of accessible "proxy targets." Project outcomes will include a scientific investigation of standard and custom psychomotor movement models for people with motor disabilities, discovery of how to rigorously measure target acquisition and text entry behavior outside the lab, the invention and validation of a simulation tool for motor-impaired performance, and the creation, validation, and deployment of a low-cost computer-based motor-ability assessment tool for use in a University of Washington assistive technology clinic.
Broader Impacts: This research will enable people with certain types of disabilities to enjoy more usable software matched to their abilities. It will also lead to a new design methodology predicated on a shift in focus from disability to ability, along with four downloadable software tools for improving laboratory, field, and clinical human motor performance evaluations. As an integral part of this research, the PI plans to organize an undergraduate summer workshop that brings informatics and computer science students together with computer users with disabilities to collectively brainstorm new access solutions. He will also develop an educational unit for middle and high school science classes that enables young students to experience simulated disabilities and learn about the roles of scientists and engineers in creating a barrier-free society.
