Introduction

The first diagnosis of dyslexia was made in 1887, but progress in the study of dyslexia was relatively slow until the educational reforms of the 1960s. Originally tackling the challenges presented to dyslexic readers by the printed page, researchers have recently begun examining the experiences of physically and, to a lesser extent, cognitively disabled Internet users. While few of these studies speak to the specific experiences of dyslexic Internet users, there is sufficient evidence to suggest that it is necessary to increase accessibility for lower-literacy users.

Dyslexia is more common than many realize. Prevalence is sometimes reported at 10%, but it has also been estimated that up to 17% of school-aged children in the U.S. experience dyslexia [1, 2]. An even greater number of people (up to 40%) possess some symptoms of dyslexia and are likely to benefit from inquiry into the dyslexic Internet users experience [2]. The effects of dyslexia are myriad and can include compromised self esteem and regard when the disability is highlighted [3]. There are laws requiring many Web sites to be accessible to those with disabilities and World Wide Web Consortium compliance is dependent on accessible Web sites [4]. Most designers understand that Web sites should be made accessible to all and that currently, further efforts are needed to extend access to disabled users [1, 5, 6]. Increasing accessibility of Web sites for people with dyslexia can also improve access for non-dyslexic users [7].

This paper seeks to uncover and explore what information about the dyslexic Internet user’s experience is already available. The scope is tightly focused on Internet usability tests of dyslexic Internet users, but selected work in parallel areas is also considered that can offer insight into the experience of dyslexic Internet users. By synthesizing existing research, it will be argued that while additional work aimed at developing a more complete profile of the dyslexic Internet user’s experience is warranted, existing research is strong enough to help guide more accessible design practices for dyslexic users. Additionally, there is sufficient evidence to argue that attempts to increase accessibility for dyslexic users can, in fact, benefit non-learning disabled users as well.

Working definition of dyslexia

The term dyslexia is used to describe a specific learning disability. For the purposes of this review, dyslexia is defined according to the National Institute of Neurological Disorders and Stroke as “a brain-based type of learning disability that specifically impairs a person’s ability to read…despite having normal intelligence” [8].

Worth noting is the fact that dyslexia may not affect learning in areas other than reading and writing and is often characterized by average intelligence. It is incorrect to pair dyslexia with attention deficit disorder or attention deficit hyperactivity disorder, as these disorders are behavioral in origin.

Dyslexia is sometimes described as a cognitive disability. In literature addressing accessibility of Web sites for people with cognitive disabilities, dyslexia is sometimes mentioned as one of the subject disabilities to be addressed [9–11] and sometimes not [12–14]. Because of the disparity in experience and ability of dyslexic readers and people with severe cognitive disabilities such as Downs Syndrome and Autism, this study of dyslexia and Internet accessibility will focus on literature devoted specifically to dyslexia rather than cognitive disabilities at large. It is believed this will yield a more specific and accurate discussion of the dyslexic Internet users’ experience.

Dyslexia is believed to have serious impact on self perception and esteem in addition to reading skills, and sometimes cause behavioral problems [3].

State of research on dyslexia and accessibility

Several studies have looked at the experiences of Web users with cognitive disabilities. Small et al. [13] concluded that current Web guidelines do not meet the needs of cognitively disabled users and called for more research into how such disabilities affect Web site use. Anderson and Rowland [14] worked with learning disability stakeholders including educators, students and parents to develop a set of technical specifications for a suite of tools to evaluate the cognitive load of Web pages. There are numerous sets of recommendations for accessible Web design that take into account the experiences of dyslexic users [9, 12, 15–19]. Most of this work studies dyslexia only as one disability within the diverse group of cognitive disabilities.

Turning to research concerned specifically with dyslexia, the bulk focuses on the neurological sources of the disability [20–22], as well as education and the development of life skills [3, 23–25]. Education materials are concerned primarily with efforts by youth to decode text while life skills materials put the onus on the reader to adapt to a non-dyslexic society. None of these approaches are greatly concerned with issues of how people interact with computers or offer research-based findings for improving Web site accessibility for dyslexic users. As this review is interested in how to accommodate dyslexic Internet users of all ages, there are shortcomings in all of these approaches.

When turning to the field of Web site accessibility, the literature addresses the two primary concerns of accommodating physical disability and cognitive disability. These explorations sometimes include dyslexia, but rarely discuss dyslexia specifically or at length. Other studies address the use of technology assistance by disabled users, but these discussions skirt the issue of usability and accessibility, instead focusing on technology implementation [26, 27].

The fields of education and accessibility often overlap, producing broad documents like “Access all areas: disability, technology, and learning” [28]. Additionally, Beacham and Alty found that different computer-based media can affect dyslexic learners’ performance, with dyslexics performing best with sound and diagram rather than textual materials [29].

Once the field of search is narrowed to the unique experience of dyslexic Web users, trends begin to emerge that suggest all users can benefit from dyslexia-friendly Web sites. Dixon’s study of students’ evaluations of a piece of education software found that dyslexic students identified more issues with greater specificity [30]. He writes that the symptoms of dyslexia are common to varying degrees among most people, suggesting a great fitness on the part of dyslexics for identifying usability problems. Usability testing of dyslexics is sparse, but a few tests have yielded results that support myriad lists of dyslexia-friendly Web style and other tools and resources aimed at extending accessibility to dyslexic users.

Usability tests of dyslexic users

There have been a handful of attempts to study Web site usability and accessibility among dyslexic readers.

In 2004, the UK’s Digital Rights Commission undertook a task-oriented examination of 100 Web sites by 50 blind, partially sighted, dyslexic, and physically and hearing impaired users [31]. Each user completed two tasks on ten Web sites. Dyslexic users experienced a 17% failure rate, which was lower than the rate experienced by blind and partially sighted users and higher than that experienced by physically and hearing impaired users. The key problems experienced by dyslexic users are common to most Internet users:

  • Confusing page layout

  • Unclear navigation

  • Poor color selections

  • Graphics and text too small

  • Complicated language

Overall, the study found that just 19% of Web sites comply even with the lowest priority checkpoints for accessibility, and all categories of disabled users consider that site designs take insufficient account of their specific need.

Al-Wabil et al. [32] conducted interviews with ten dyslexic Web users and concluded that there are considerable barriers to Web use for dyslexics. This exploratory study revealed several patterns of behavior among the interviewees, including frustrations with both Web site structures and textual presentations. Responses to structural features such as navigation trails varied based on the severity of reading impairment, with less impaired users stating that navigation trails helped them understand where they were within a site, and moderately impaired users stating that navigation trails were helpful for backtracking, but not for keeping track of their location within a site. Other patterns surrounding use of search functions and back and forward browser buttons were also identified.

Kurniawan and Conroy [33] tested dyslexic and non-dyslexic students for reading comprehension speed and accuracy. Participants read materials of varying complexity and the authors concluded that dyslexic participants made increasingly frequent mistakes as reading material became more complex. This study also investigated the technological aids employed by dyslexic Internet users, such as screen readers and colored overlays, and found that allowing users to select their optimal color scheme increased reading speed for both dyslexic and non-dyslexic readers.

The DIAMONDS project was a study of ten Internet users’ experiences with a piece of tutorial software [1]. The study was aimed at determining how best to move from accessibility awareness to practice. Of the ten users, nine were dyslexic and one was dyspraxic, having trouble executing fine motor tasks. The study used a combination of interviews, questionnaires, log sheets, and a focus group. Observation and talk-aloud protocol analysis were considered, but the software being tested required the students spend time at home working independently. Results indicated a desire among dyslexic students to customize the fonts, type size and colors of on-screen text. Due to the methodological objective of the study, full results of the usability test were not published.

In addition to the call for highly configurable text for dyslexic users, simplifying language and organization is also common. Jakob Nielsen [34] found that lower-literacy readers experience Web texts very differently than high literacy readers. The study cited the US Department of Education’s National Adult Literacy Survey, which asserted that 48% of the U.S. population has low literacy, and Nielsen’s estimate that 30% of Web users have low literacy. The study tested regular and low literacy optimized versions of a Web site with lower and higher literacy users. The three metrics included success percentage, task time and satisfaction. Recommendations for optimization included: prioritize information, avoid text that moves or changes, streamline the page design, simplify navigation, and optimize search. Nielsen found that the Web site optimized for low literacy readers resulted in improvements in all metrics: percentage, task time and satisfaction. The improvements for lower literacy readers also appeared to improve, rather than harm, the experience of high literacy users, as well.

Recently, Swierenga et al. [35] conducted a usability evaluation of the Michgan.gov voting Web site with users who have dyslexia. The usability evaluation of the voting information on the Michigan.gov Web site involved six participants who have dyslexia, specifically college students of voting age. Participants were asked to perform typical tasks such as locating voter registration requirements and ballot issues, and their activities were recorded (audio and videotaped). Participants were then asked to comment on their experience with the site and offer recommendations they believed would have made the site more usable. The results indicated several problems with the user interface for dyslexic users that the authors used to develop a list of recommended practices for more universal design.

These studies tell a great deal about dyslexia and accessibility. Many Web sites do not meet established accessibility standards, a situation that poses a significant barrier for many Internet users. It is believed that nearly one-third of Internet users are either dyslexic or possess some symptoms of dyslexia, and that the failure to address accessibility for dyslexics likely negatively affects their experience as users. Research-based recommendations are available for improving accessibility for dyslexics, such as allowing users to control the presentation of text. Perhaps most compelling, it is also well known that adjusting Web texts to make them more accessible to lower-literacy users also improves the experience of higher-literacy users as well.

How multimodal documents help dyslexic users

Research commonly notes that dyslexia is highly variable; there is no “typical” dyslexic Internet user. The results of Pollak’s interviews with dyslexic college students underscore the potential strengths of multimodal documents for differently abled readers, rather than the “one document for all” approach [25]. Some researchers have addressed this idea, developing and testing highly configurable software.

SeeWord is a word processing software that allows users greater control over how information is displayed [36]. In response to an unpublished pilot study of dyslexic writers and computers that found there was no universal profile of dyslexics, SeeWord was designed to allow users to optimize writing and reading conditions in Microsoft Word for their own needs. The evaluation of SeeWord employed a “talk aloud” task-oriented observation, and concluded that some dyslexic users read more accurately when they are able to select their own text color and size settings.

Researchers have addressed readers’ diverse abilities in other software development as well. The MultiReader system marks an attempt to create rich multimedia communications that are highly configurable for both user needs and preferences [6]. The project rejected the “one document for all” approach to accessibility, instead synchronizing multiple media types and affording the user more control over presentation. The project included an evaluation of the MultiReader system with 70 print-disabled users, and found that enabling greater control over presentation and synchronization of screen reader audio and text highlighting benefited dyslexic users. Petrie concludes by pointing out that adequate tools are in place for blind users, but that deaf and dyslexic users require more effort on the part of accessibility professionals.

Guidelines and resources

Despite the scarcity of research literature addressing Web accessibility for dyslexic users specifically, efforts to make more effective human–computer interactions for dyslexic and other disabled users are being made. Nielsen [34] recommends making text size selectable, using sans serif fonts, and using colors to visually designate important pieces of information. These measures are also recommended for dyslexic readers [2]. Many dyslexic Internet users employ screen readers for assistance, and so efforts to optimize Web sites for screen readers may benefit dyslexic users, even if the stated target audience is blind users [26].

A typeface has been developed specifically to address the recognition difficulties faced by dyslexic users. Dyslexic-friendly typefaces for digital environments have been limited due to the challenges of creating a highly pixilated sans serif that differentiates between similar letters such as “b” and “d”. The Read Regular typeface more dramatically changes the profile of such letters and has been found to ease computer based reading for dyslexic users [37].

There are a host of recommendations aimed at developing Web sites friendly to dyslexic users.

  • Pearson [38] outlines steps for designing online courses that include dyslexic readers.

  • Rainger [16] synthesizes practitioners’ experiences and usability research to illuminate the issues of accessibility and dyslexia.

  • Bradford and Zarach [18, 39] offer style guidelines for making Web sites accessible to dyslexics.

  • Jiwnani [17] makes recommendations for designing Web sites that are easier to read both by the cognitively disabled and the assistive technologies that the cognitively disabled often employ.

  • Online tests can provide Web authors with immediate readability evaluations based on the Gunning Fog, Flesch-Kincaid and other reading level algorithms [40], but the accuracy of these tests has been questioned [28].

  • Phipps et al. [28] offers guidelines for dyslexia friendly interfaces within the comprehensive TechDis book, “Access all areas: disability, technology, and learning”.

Most of the guidelines listed above (and those employed by Nielsen in his study of low-literacy users) are also included in guidelines for improving access for people with other disabilities. In fact, Evett and Brown compared and drew contrast of overlap between recommendations for different disabilities [41]. That study, which concerned itself with the text and Web style guidelines for the Royal National Institute for the Blind and British Dyslexia Association, concluded that recommendations for blind readers closely parallel those for dyslexic readers. It also found that non-disabled users benefit greatly from adherence to accessibility guidelines aimed at blind, partially sighted, and dyslexic users.

Plain language/minimalism

The recommendations for designing dyslexia inclusive Web sites outlined above are not exclusive to the field of learning disability. As several studies already discussed have found, guidelines for dyslexia friendly Web sites can improve usability for non-dyslexic readers, too [30, 33, 34, 41]. Many of these guidelines are already accepted tenets of plain language, the movement for clear, concise, jargon-free texts. Plain language and accessibility have been linked in the past [7].

Originally a consumer response to legalese in business and government documents, several states have passed or considered plain language laws [42]. The argument has been made that plain language is a vital part of Web accessibility and that its benefits are shared by disabled and non-disabled users alike [7]. This suggests that a study of dyslexia and Web accessibility could increase the accessibility of Web documents for all users, regardless of ability.

Conclusion

While great efforts have been made to research both dyslexia and Web accessibility, there is not a great body of work addressing both simultaneously. The majority of accessibility efforts are aimed at blind and visually impaired users or, to a lesser extent, users with severe cognitive disabilities. Within and adjacent to these studies, though, is a small but compelling body of work that, once uncovered, calls out clearly for increased attention to dyslexia and accessibility.

Meanwhile, evidence mounts that the Web sites that eschew guidelines for dyslexic accessibility not only undercut the esteem and successes of dyslexic users, but also their non-dyslexic counterparts. The high level of overlap in guidelines for dyslexia, other cognitive and physical disabilities, and plain language, suggest that dyslexic users could be a salient indicator group for overall Web accessibility. By undertaking more task-oriented usability research employing observation and talk-aloud protocols, a profile of the dyslexic Internet user’s experience can be established. Such a profile would confirm or disconfirm the assumptions already commonly held, and strengthen the basis for accessibility initiatives that benefit users with and without dyslexia. Even without the confirmation of such additional studies, though, accumulated knowledge of dyslexia and accessibility offers tools to improve the dyslexic users experience and is sufficient to argue that access for dyslexic users is in fact a matter of access for everyone.