Designing for Users With Cognitive Disabilities

Abstract

Users with cognitive disabilities interact with technology in different forms. Designers need to understand the deficits of users with cognitive disabilities in order to design materials that are accessible to those users. This paper provides an overview and analysis of the current state of service to those with cognitive disabilities, and makes practical suggestions on design issues, as well as suggesting further areas for research.

Introduction

Users with cognitive disabilities include users with learning disabilities with or without intelligence deficits, and those with language disorders, including Alzheimer's patients, stroke victims and those with other forms of brain damage. These users can be of any age or economic group, although learning issues tend to be addressed primarily for younger users, and those with language disorders are often elderly. A great deal of work has been done on service for the physically disabled, and that work will not be duplicated here. For issues related to assistive technology designed to accommodate physical needs, the reader is referred to http://www.acm.org/sigcaph/links. This site has many links focused on accessibility resources for the physically handicapped.

Although this discussion will be limited to cognitive disabilities, this still covers a wide range of issues. Learning disabilities range from reading issues (dyslexia), to writing issues (dysgraphia), to computational disabilities, trouble with numbers and other areas that relate to numerical manipulation. Frequently, students with learning disabilities have trouble with abstract reasoning and organization. Sometimes, learning disabilities go hand-in-hand with Attention Deficit Disorder and Hyperactivity (ADD or ADHD). Victims of language disorders may lose their ability to communicate effectively and have short and long term memory loss. Language difficulties may include word finding and syntax development difficulties. Sufferers of language disorders may have lost the ability to organize, assimilate, and retain information. Often, these cognitive disabilities are accompanied by physical disabilities.

The current state of literature and research is introduced in the Background section. This literature is discussed in further depth in the Analysis section. The Guidelines section makes practical suggestions for designing interfaces for users with cognitive disabilities. The Conclusion suggests areas for further research and study.

Background

The current literature is focused in three main areas: surveying the current theory on the use of technology for a particular disability; reporting on experiments and case studies; and using and adapting the current technology for specific purposes. Most of what has appeared discusses aspects of the three main areas in which technology has been historically used for the cognitively disabled. These are:

• diagnostics and assessment
• remediation and instruction (Computer Assisted Instruction -- CAI)
• compensation (assistive technologies)

Computer Assisted Instruction ranges from software designed to provide remediation in specific areas through repeated drills, to multi-media based, open-ended packages designed to encourage the development of written language. Studies and experiments have been done on the use of these systems in the classroom, and with language disorders [13, 23]. This work is primarily focused on the theory of education, although there are practical papers providing assistance for teachers and parents and users [2].

Providing compensation and/or assistive technologies for cognitive difficulties is a significant area of discussion and research. This is particularly true because the law mandates these accommodations. The Americans with Disabilities Act of 1990 (ADA) provided that reasonable accommodations need to be made for individuals with disabilities in the workplace. The Individuals with Disabilities Education Act (IDEA) and its 1997 amendments provided that assistive technology devices and services are required to be made available to students and youth with disabilities if these are necessary to ensure a free, appropriate public education. A number of studies have been done on different assistive technologies, and their efficacy in serving cognitively disabled users. Research has been done in the use of word processing systems for all age groups [12, 17]. The discussions include the identification of the most useful aspects of word processing, including grammar and spelling checks, outlining, and brainstorming assistance, and speech synthesis and text read-back. Work has been done in the use of speech recognition software for writing, [7] and managing technology [12]. Authors also include personal data managers and organizational tools in their discussions [17], as well as using technology for information access and fostering special talents and abilities in those with cognitive disabilities [19].

Analysis

Most current literature focuses on how tools that currently exist are being used. In the case of CAI, researchers have primarily examined the efficacy of the method, rather than the efficacy of the design. Some research does look at why some software is more effective than others. Larsen [10] identifies difficulties with distracting interfaces and animation in remediation software. Others [3] identify problems with cluttered interfaces causing distraction to learning disabled users. They also mention the importance of matching the remediation method chosen to in-class educational methods and the importance of creating modifiable programs that can be tailored to a student's needs while providing frequent and informative feedback.

There is limited research on the efficacy of some assistive technologies. However, in [8] the authors discuss a program at California State University, Northridge (CSUN) which is transferring research results from experiments on assistive technologies into practical applications for learning disabled students, and successfully measuring the results. Most other studies are limited or papers are based on anecdotal evidence. Some of the previous research on word processing has become outdated as technology has changed. New studies point to the development of graphical interfaces, the use of pointing devices including the mouse, and speech synthesis as felicitous changes for learning disabled users, but again, most research is in the form of case studies [17, 18].

In a discussion on speech recognition (SR), [7] the author notes the advantages and disadvantages that may accrue to learning disabled users from using SR in producing written work. In particular she addresses the trade-off between removing the physical barrier of the keyboard versus the cognitive load of oral composition. She also discusses the steep learning curve necessary to operate the system. In her conclusions, she recommends that users should build their "recognition templates" (i.e. train the system) in settings where the system will be used. She also discusses the necessity for consistency of language, practice, and training with commands and development of error correction procedures. Her implication is that users would need to be carefully and extensively trained on current technology, but makes no suggestions for changes that could/should be made in future technologies.

In cases where technology development is discussed, a differentiation needs to be made between designing for specific software, and designing for the World Wide Web. Although suggestions may be effective in both areas, some of the suggestions made are contradictory with accepted web development policy. In the areas of software development for learning disabled students, authors point to reducing clutter on the screen, consistency of design, and modifiability [2,15]. Web users with learning disabilities have issues with graphic pages that cannot be accessed using text readers and confusing and inconsistent content [25]. In a discussion of issues that affect people with language disorders, [21] the overwhelming amounts of text on web pages, and messages from the computer (the ubiquitous "404 - Page not found") are identified as areas that can cause confusion and panic. However, some suggestions for improving access by those with cognitive disabilities contradict modern web development theory. For those with cognitive disabilities, it is better to place less information and fewer links on a page, suggesting deep narrow search trees. This contradicts the current development practice of having shallow wide search trees.

Guidelines and Recommendations

Recommendations that benefit users with cognitive disabilities can also benefit the population at large. For example, multiple methods of access benefit everyone. To improve the ability to access the web using different methods, designers should follow the "Web Content Accessibility Guidelines 1.0" released by the World Wide Web Consortium (W3C) http://www.w3.org/WAI/. Designers should also include people with cognitive disabilities in the design, research, and development of new technologies [25], and they should be sure to create designs that are inclusive across cultures, and not jingoistic [19]. All designers should remember Shneiderman's 8 golden rules of interface design [20]. In particular, users with disabilities benefit from designs that stress consistency, informative feedback, error prevention with friendly error handling, and reduce short-term memory load.

The primary focus for designers producing any material for those with cognitive disabilities, whether it be new computer aided instruction, assistive technologies, or the World Wide Web should be to:

• Avoid clutter
• Create modifiable software
• Provide software with useful, informative, and frequent feedback
• Provide multiple methods for access

In designing for the web specifically, designers need to balance the demands of many different types of users. It might be efficacious to provide different access methods. Although graphics are becoming more ubiquitous, many sites offer a text-only option. It is important, though, to update the text-only site in parallel with the graphic site, and not allow the content on the text-only site to become dated.

Designers must remember to provide:

• Support for text browsers on the WWW
• Alternative methods for on-line forms
• Information layouts that are consistent and easy to understand. For example, display important information in a prominent area to catch the user's eye
• Simplified and consistent design and presentation

Designers of graphical interfaces who are considering the effect of conversion from video to audio should also see Media Conversion from Visual to Audio Voice Browsers included in this website.

Developers might also want to consider the following suggestions made by Singh, Gedeon, and Rho (1998). They suggest that designers should:

• Remove the need for reading and writing skills in web exploration through graphic representations and point and click interfaces.
• Create search techniques that are easy to use, and make them consistent across sites, providing as many graphical and point and click options as possible.
• Reduce information overload, by simplifying text or providing options for abbreviated content.
• Develop specialized software agents to assist language-impaired and learning disabled users with searches
• Provide useful ranking systems for search engine results
• Provide the opportunity for audio and video representations to cue subject recognition.

Conclusion

"In the past decade, technology has become smaller, cheaper, more powerful, and easier to use, and these trends are likely to continue." [12] Three changes have contributed to increased use. These are: the move away from command-driven control to graphical interface; the standardization of interfaces; and the provision of input alternatives. As technology continues to change, the adaptation of that technology to disabled users must parallel the changes. The efficacy (or lack thereof) of assistive technologies for those with cognitive difficulties needs to be researched and documented. Then, researchers need to ensure that their work will ultimately benefit the user. It is necessary to create mechanisms that transfer validated research to actual implementation [17].

Research also needs to be done to test today's current theories when applied to users with disabilities. The depth versus breadth question for hierarchical structures needs to be re-examined. This may in fact be a case of what is good for the 'abled' not being good for the disabled. The question of information overload also needs more research to determine how content can be simplified without losing information. Providing services to the disabled should not impose restrictions on the non-disabled, and vice versa. It is important that design decisions are tested on diverse groups of users to determine if all are being successfully served.