Shari Trewin

Effect of age and Parkinson's disease on cursor positioning using a mouse

Point-and-click tasks are known to present difficulties to users with physical impairments, particularly motor- or vision-based, and to older adults. This paper presents the results of a study to quantify and understand the effects of age and impairment on the ability to perform such tasks. Results from four separate user groups are presented and compared using metrics that describe the features of the movements made. Distinct differences in behaviour between all of the user groups are observed and the reasons for those differences are discussed.

Developing steady clicks:: a method of cursor assistance for people with motor impairments

Slipping while clicking and accidental clicks are a source of errors for mouse users with motor impairments. The Steady Clicks assistance feature suppresses these errors by freezing the cursor during mouse clicks, preventing overlapping button presses and suppressing clicks made while the mouse is moving at a high velocity. Evaluation with eleven target users found that Steady Clicks enabled participants to select targets using significantly fewer attempts. Overall task performance times were significantly improved for the five participants with the highest slip rates. Blocking of overlapping and high velocity clicks also shows promise as an error filter. Nine participants preferred Steady Clicks to the unassisted condition. If used in conjunction with existing techniques for cursor positioning, all of the major sources of clicking errors observed in empirical studies would be addressed, enabling faster and more effective mouse use for those who currently struggle with the standard mouse.

Biometric authentication on a mobile device: a study of user effort, error and task disruption

We examine three biometric authentication modalities -- voice, face and gesture -- as well as password entry, on a mobile device, to explore the relative demands on user time, effort, error and task disruption. Our laboratory study provided observations of user actions, strategies, and reactions to the authentication methods. Face and voice biometrics conditions were faster than password entry. Speaking a PIN was the fastest for biometric sample entry, but short-term memory recall was better in the face verification condition. None of the authentication conditions were considered very usable. In conditions that combined two biometric entry methods, the time to acquire the biometric samples was shorter than if acquired separately but they were very unpopular and had high memory task error rates. These quantitative results demonstrate cognitive and motor differences between biometric authentication modalities, and inform policy decisions in selecting authentication methods.

Abstract user interface representations: how well do they support universal access?

This paper examines four XML languages for abstract user interface representation: UIML, XIML, XForms and AIAP. It discusses whether the high level architectures of these languages support the requirements of universal usability by allowing use of personal interfaces. Specific technical requirements include separation of data from presentation, explicit declarative representation of interface elements, their state, dependencies, and semantics, flexibility in inclusion of alternative resources and support for remote control and different interaction styles. Of the languages examined, XForms and AIAP provide the best match to the requirements. While XForms requires an appropriate delivery context to provide full access, the AIAP standard will include specification of the context in which the language is to be used.

Automating accessibility: the dynamic keyboard

People with motor disabilities may need to adjust the configuration of their input devices, but often find this an obscure and difficult process. The Dynamic Keyboard exemplifies a potential solution. It continuously adjusts fundamental keyboard accessibility features to suit the requirements of the current user, based on a keyboard use model. Early field results indicate that users have not chosen to take control of these accessibility features from the Dynamic Keyboard, and that a variety of settings are being used. A more detailed ongoing study suggests that automatic adjustment of the key repeat delay feature is acceptable to users, while the debounce feature may not be appropriate for dynamic adjustment.

Improving web accessibility through an enhanced open-source browser

The accessibilityWorks project provides software enhancements to the MozillaTM, Web browser and allows users to control their browsing environment. Although Web accessibility standards specify markup that must be incorporated for Web pages to be accessible, these standards do not ensure a good experience for all Web users. This paper discusses user controls that facilitate a number of adaptations that can greatly increase the usability of Web pages for a diverse population of users. In addition to transformations that change page presentation, innovations are discussed that enable mouse and keyboard input correction as well as vision-based control for users unable to use their hands for computer input.

Automatically detecting pointing performance

Since not all persons interact with computer systems in the same way, computer systems should not interact with all individuals in the same way. This paper presents a significant step in automatically detecting characteristics of persons with a wide range of abilities based on observing their user input events. Three datasets are used to build learned statistical models on pointing data collected in a laboratory setting from individuals with varying ability to use computer pointing devices. The first dataset is used to distinguish between pointing behaviors from individuals with pointing problems vs. individuals without with 92.7% accuracy. The second is used to distinguish between pointing data from Young Adults and Adults vs. Older Adults vs. individuals with Parkinsons Disease with 91.6% accuracy. The final data set is used to predict the need for a specific adaptation based on a users performance with 94.4% accuracy. These results suggest that it may be feasible to use such models to automatically identify computer users who would benefit from accessibility tools, and to even make specific tool recommendations.

PowerUp: an accessible virtual world

PowerUp is a multi-player virtual world educational game with a broad set of accessibility features built in. This paper considers what features are necessary to make virtual worlds usable by individuals with a range of perceptual, physical, and cognitive disabilities. The accessibility features were included in the PowerUp game and validated, to date, with blind and partially sighted users. These features include in-world navigation and orientation tools, font customization, self-voicing text-to-speech output, and keyboard-only and mouse-only navigation. We discuss user requirements gathering, the validation study, and further work needed.

Physical usability and the mobile web

This paper examines the degree of overlap between good design for physical ease of access on the Web in general, and design for physical ease of use on the mobile Web. There are marked differences in the basic interaction techniques used and usability issues experienced. As a group, people with physical impairments tend to have a broader range of needs. These differences impact Web page design in various ways. Problems can be addressed in a unified way by designing for device independence. At least for physical ease of access, a unified set of mobile/accessibility best practice guidelines would be mutually beneficial. This approach may be helpful in preventing fragmentation of the Web.

Configuration agents, control and privacy

When a truly diverse community of users is considered, it is clear that a single interface, no matter how well designed, cannot accommodate everyone. Configuring an interface to match a particular users requirements can be a demanding task. This paper focuses on the potential utility of configuration agents to support users with low initial levels of control, or low levels of knowledge about configuration. The relationship between configuration of systems by agents, user control, and user privacy is discussed. It is argued that privacy issues are of primary importance, and suggestions are made for ways to apply agent technologies to the configuration problem without compromising user control and privacy. Work in progress to inform and further develop these issues is described.