Significant barriers exist for individuals with intellectual disability to independently access print-based content. It is regrettable that, while the amount of content now available electronically increases, tools to access these materials have not been developed with individuals with intellectual disability in mind. This article reports the results of research evaluating the use of a palmtop PC-based application designed to enable individuals with intellectual disability to access electronic books and documents. Participants with intellectual disability were randomly assigned to 1 of 3 groups, each group differing in the type of audio player used. Participants who used the specially designed reader made significantly fewer errors accessing electronic books and required significantly fewer prompts than did participants using either of 2 mainstream audiobook readers.
Research findings document the potential benefits of computer use by people with intellectual disability for a wide array of outcomes (Wehmeyer, Smith, Palmer, Davies, & Stock, 2004). For example, people with intellectual disability have used computers to acquire skills needed to perform vocational tasks (Riffle et al., 2005; Wehmeyer et al., 2006), as augmentative communication devices (Hetzroni, Rubin, & Konkol, 2002), to learn functional tasks (Lancioni, O'Reilly, Campodonico, & Mantini, 2001), to engage in antivictimization training and language-acquisition activities (Holzberg, 1994), for menu planning (Stock, Davies, & Ocken, 2001), for independent Internet access (Davies, Stock, & Wehmeyer, 2001), and for decision-making support (Davies, Stock, & Wehmeyer, 2003).
One of the most frequently identified barriers to greater independence and integration for people with intellectual disability involves accessing written materials and content. Most people with intellectual disability have difficulty reading. Instruction to promote literacy for this population has predominantly focused on teaching sight-word vocabulary skills (Browder, Courtade-Little, Wakeman, & Rickelman, 2006). Browder et al. reviewed 128 studies focusing on literacy for persons with intellectual disability from 1975 to 2003 and found that 80 of these targeted sight-word instruction. Fewer (n = 36) focused on fluency and fewer still focused on comprehension (n = 31), phonics (n = 13), or phonemic awareness (n = 5; Browder et al., 2006, p. 69). Although initiatives such as the U.S. Department of Education's Reading First initiative (Elementary and Secondary Education Act [ESEA], 1965) and federal policy detailed by the No Child Left Behind Act (NCLB; 2002) and the Individuals With Disabilities Education Act (IDEA; 1975) have resulted in a renewed interest in literacy instruction for students with intellectual disability— efforts that may provide strategies to achieve greater progress in literacy for people with intellectual disability—it seems likely that many people with intellectual disability will need additional supports to fully access written or text content and use such information to achieve employment, community inclusion, and recreation goals. Providing technology-focused supports and electronic or digital text to enable people with intellectual disability to access written content has the potential to accomplish these goals.
Providing access to written content in electronic or digital formats, including audiobooks, textbooks, and other digital materials, holds considerable promise in giving people with intellectual disability the opportunity to engage with print-based information. Electronic or audio versions of a wide array of print materials currently exist. For example, Recordings for the Blind and Dyslexic (2006) has reported the availability of over 139,000 books in audio format, many of which are now available in digital formats for download. Another example is provided by netLibrary, a Web-based eBook (eBooks may include virtually any text that is in electronic format) repository that includes over 40,000 eBooks. netLibrary reported in 2004 that there were over 54,000 eBook titles available.
eBooks are currently available as either raw text or binary file format documents. Raw-text eBooks are available throughout the Internet. However, the availability of books in this format is diminishing, as many of these books are being replaced by documents in binary formats. Binary file format eBooks appear to be a much more popular format than raw-text books, and this is very likely due to the security and piracy issues with copyrighted material, as well as the portability of having an entire book encapsulated in a single file versus multiple HTML pages. There are several binary formats currently in use, including DigitalWebBook and OEB Package Format, but the most common are Microsoft's .LIT format and Audible's .AA format. Microsoft, Adobe, and Audible offer eBook reader software and methods for purchasing and downloading copyrighted material through the reader software. There are numerous websites available to download and purchase eBooks, and many large book distributors, such as Amazon.com, dedicate entire sections of their website to eBook purchases. In fact, many libraries now provide free downloads and time-limited use of current books in eBook formats.
The leading eBook reader applications have some accessibility features designed primarily for individuals with physical and visual disabilities, but, as is often the case, these do not include features designed to make eBooks accessible to individuals with intellectual disability. The interface for selecting and listening to an eBook with these applications requires the ability to read, as well as the ability to navigate through text-based menus for selecting books. In addition, managing playback and navigation of the selected materials is prohibitively complex for individuals with intellectual disability, given existing user interfaces.
In this study, we investigated the utility of a simplified, multimedia, palmtop computer application to enable individuals with intellectual disability to independently access eBooks and documents by listening to recorded or computer-generated speech.
Study participants included adolescents with intellectual disability recruited from a local education agency's transition program and adults with intellectual disability supported by multiple community-based agencies. All participants were receiving community-based services and supports across a variety of functional life–activity areas, including employment, academic, and daily living skills. A total of 49 participants were involved in the study, 17 of whom were female and 32 male. The average age of participants was 28.47 years (SD = 10.35), with ages ranging from 18 to 60 years. The average IQ score for participants was 52.88 (SD = 8.69), with a range of 31 to 69. No specific measures of previous comfort level, familiarity, or anxiety using computer or technical devices were recorded. Thus, participants were randomly assigned to one of three conditions identified by the type of electronic book reader program they were to use in the study, described subsequently. Table 1 summarizes age and IQ data for participants in each of the three conditions.
Informed consent was obtained from all study participants prior to the start of the study in accordance with the approved Institutional Review Board procedures for this project. Participants were paid a stipend for their involvement.
The software used in this study was a cognitively accessible software program, called Rocket Reader (AbleLink Technologies, 2006), designed to enable access to electronic books for individuals with intellectual disability. The reading software is operated on the Pocket PC palmtop computer platform. The user interface is highly visual and auditory based and provides consistent feedback and interaction with the user to maximize usability and simplicity. The landscape mode that is available in the Pocket PC 2003 operating system was used for the system, as this orientation provided a more natural interface for holding the unit and navigating through the books with the left and right thumb as well as for displaying two book images large enough to be recognizable to the user. All hardware buttons on the Pocket PC are automatically disabled to prevent accidental activation of other programs. When the program first starts, the first two books in the library are presented (in alphabetical order) and an audio prompt instructs the user to select the book to which he or she would like to listen. Additional books are accessed by using arrow buttons on the bottom left and right of the display. These arrows are only visible if the library contains more than two books.
One common barrier to accessibility inherent in mainstream Pocket PC programs involves the use of very small buttons and controls that require precise click actions with a stylus. The cognitively accessible reading system was intentionally designed not to require use of a stylus but to allow users to operate the touch screen with their fingers if they chose. For example, in the program, the entire circular area surrounding the navigation arrows is the active area, so that these buttons can be easily tapped with a finger or thumb.
When the user taps on the picture of the book he or she wishes to listen to, the book is highlighted (e.g., Charlotte's Web [White, 1952]) the title and author are displayed, and an audio prompt is played announcing the title of the book, such as “Charlotte's Web—to select this book, touch this picture again.” Although text is displayed onscreen, users with limited or no reading skills are still able to identify a desired book using the “Button Talk” feature that verbally announces the name of the book when the icon is first tapped. The book is not actually opened until the user taps the book icon a second time to select it. Therefore, nonreaders can sequentially tap each book image and listen to the name until they find the book they are seeking. The audio prompt then instructs users to tap the book a second time if they would like to open that book and begin listening.
The audio prompt that plays when the book is tapped can be customized, including audiorecorded by a parent or caregiver, so it can be completely customized to the needs of the user. For example, an audio prompt could say, “Click on this picture again to read your favorite story.” Once the user locates the book he or she is looking for and taps it to hear the name of the book, tapping a second time opens the book and begins an audio output of the book. After the desired book is selected, the screen changed to depicts the cover of the book and begins reading the book. If this is the first time the book has been opened, the audio is started at the beginning of the book. If the book has been listened to previously, the system opens the book and picks up the audio presentation where the user left off the last time. For feedback purposes, an animated icon of a speaker is displayed on the left side of the screen to denote that the book is currently being played. Below the speaker icon is a progress meter that shows the relative location in the book. On the right side of the screen is the book cover. The available buttons on the screen, from left to right, are an orange Restart button, a red Stop button, and a Read/Pause button (green and blue, respectively). The Restart button allows the book to be restarted from the beginning. The Stop button stops the book and immediately returns to the book-selection screen, and Read/ Pause allows the book to be paused and played again and changes appearance based on the current play state.
The user's current location within a book is always saved on the Pocket PC. If the user turns off the Pocket PC while the book is playing and turns it on again later, the book will resume where it left off. If he or she stops the book, the location is saved, and, when the user returns to the book, it will resume a few seconds before the saved point, to lead the listener back into the story.
In summary, the cognitively accessible audioplayer was designed for use by individuals with intellectual disability and provides a simplified interface; a limited number of large, easily selectable buttons; and both visual and audio cues and instructions to help the user operate the system. The system includes a number of user-customization options that allow the interface to be customized according to the needs of the user. For example, for users who want more complexity, the personal library can be organized into categories, with a picture or icon representing the category (e.g., mystery books using a magnifying glass for an icon) that, when selected, can display as many as eight book covers on the display at a time, with the navigation arrows then taking the user to more books in the selected category. Alternatively, for users who desire less complexity, another interface option that is even simpler displays only one book image at a time, and each time the navigation arrow is pressed, the next book is displayed and the name of the book is announced with an audio message. This approach provides the simplest operation for selecting books and can also be used by listeners with visual impairments who need to step through the books one at a time and hear the title before selecting a book.
The study compared the degree to which participants were able to independently and accurately utilize the audio reader—designed specifically for use with people with intellectual disability (Rocket Reader)—with similar indicators with two other commercially available audioplayers: the Audible Player (audible.com, Newark, NJ) and Microsoft Reader (Microsoft, Redmond, WA). The Audible Player is designed to play digital files in multiple formats, including eBooks. Microsoft Reader is the eBook player designed for use on Pocket PCs and is often bundled with the Windows Pocket PC operating system. Participants were randomly assigned to one of the three groups, as determined by which audioplayer was utilized, with 17 participants in the Rocket Reader group and 16 each in the Audible Player and Microsoft Reader groups.
The study took place over a 4-week period. Data were collected on two dependent measures of utility: accuracy and independence. Accuracy was defined as the person's ability to select a book and operate the system using the playback procedures, and independence referred to the number of prompts a participant needed to complete a testing session. In addition, there were two equivalent book-selection scripts that were randomly selected for use during the person's test session. Each player was loaded with the same eight audiobooks. Each script required users to select four different books to listen to, to close each book when instructed, to return to the book-selection screen, and to return to the books and continue reading at the previous location. The Appendix provides one of the two book-selection scripts, with specific verbal instructions that were provided and associated timing for presentation of the instructions. The second book-selection script was identical, except for the specific books that that user was directed to open and listen to. Last, the three player applications were presented using two of three different Pocket PCs: a Dell Axim X50, a Toshiba E-400, and an Asus MyPal Pocket PC (e.g., each audioplayer was presented to participants on at least two different Pocket PC devices). Each Pocket PC ran the Pocket PC 2003 operating system.
Prior to beginning the book-selection scenario, a training script was used to illustrate correct task performance by first demonstrating book selection and playback to the participant. This demonstration was followed by having participants use the player application to select and play back a book on their own and then to return to a bookmarked location to pick up listening where they left off. This training activity was designed to teach participants how to operate the specific player for his or her experimental condition. Each training session was conducted individually for each participant immediately prior to the start of the individual's experimental session. Following this training period, the participant was requested to use the player to select a series of books and to begin listening to them. The two book-selection scripts required the user to select a book, begin listening for a period of time, and then close the book and select a new book. Four different books were opened during the testing period, and three books were required to be opened twice so that the user could demonstrate his or her ability to return to a book that had been opened previously and continue listening from where they left off. Each script required a total of 12 separate book-selection tasks (see Appendix). The duration of test sessions for each participant ranged from approximately 15–30 min, depending on the person's speed at performing the tasks.
Test Procedures and Experimental Variables
Data collection forms were used to record errors and prompts during each experimental session. An error was recorded (a) if the user did not perform the step correctly, (b) if the user skipped the step completely, or (c) if the user initiated an action that did not have the intended result, such as attempting to tap a button but missing it. A prompt was recorded (a) if the user specifically requested help or (b) if a prompt was necessary to correct an error prior to proceeding to the next step.
Two data collectors scored all the participants in parallel to allow for interrater reliability measures to be obtained. Interrater reliability of ratings for recorded errors was 98% and for recorded prompts was 96%. The ratings made by the two raters were averaged to then provide a single error and prompt score for each participant for data analysis. In addition, there was room provided on the data collection forms to record additional observations as well as statements made by participants during the test sessions.
A one-way analysis of variance (ANOVA) procedure was conducted to evaluate mean differences between the three experimental conditions (Rocket Reader, Audible Player, and Microsoft Reader) for each dependent measure: mean errors and mean prompts. Data were analyzed using SPSS 13.0.
Table 2 provides means and standard deviations by experimental condition and dependent measure. There were significant differences on univariate ANOVA tests between groups for both average number of errors, F(2, 46) = 6.725, p < .003, and average number of prompts, F(2, 46) = 5.817, p < .006. Table 3 summarizes the results of the ANOVA tests. Results from post hoc analyses, using Tukey's honestly significant difference test, are provided in Table 4. For average errors, which was a measure of participants' ability to correctly operate the three different player applications, pairwise comparisons showed no differences when participants used the Audible Player (M = 12.59, SD = 12.92) compared with when participants used Microsoft Reader (M = 14.15, SD = 13.33). However, when using Rocket Reader (M = 1.67, SD = 2.39), participants made significantly fewer errors compared with the Audible Player (p = .014) as well as significantly fewer errors compared with the Microsoft Reader (p = .005).
For the average number of prompts provided to participants while performing the book-selection tasks during the experimental sessions, pairwise comparisons showed no differences when participants used the Audible Player (M = 13.56, SD = 12.51) compared with when participants used the Microsoft Reader (M = 14.65, SD = 12.95). However, when using Rocket Reader (M = 3.26, SD = 4.54), participants required significantly fewer prompts to complete the book-selection and playback tasks compared with using the Audible Player (p = .021) as well as significantly fewer prompts compared with using Microsoft Reader (p = .010).
The findings from this study demonstrate that an audio eBook reader designed with features that promote cognitive access can be used by people with intellectual disability with greater accuracy and increased independence than standard, commercially available, eBook audio readers. When using Rocket Reader, 76% of study participants with intellectual disability were able to complete the 12-step book-selection scenario with one or fewer errors. When using the Audible Player, only 25% of participants were able to access the book-selection scenario with one or fewer errors, and, similarly, when using Microsoft Reader, 25% were able to access the book-selection scenario with one or fewer errors.
There are a number of limitations that must be taken into account when considering these results. First, although participants were randomly assigned to groups that differed according to the audio reader software used, the size of each group was relatively small and replications are needed with larger samples. Second, because of the limited number of Palmtop PCs with which to conduct the study, it was necessary to have participants come to a central location and use the device there. This provided a relatively short time during which the software use could be evaluated, and, as such, we were not able to examine more than basic device use features. As such, this study focused on limited variables (i.e., ease of use, accuracy) and did not address impact on variables such as motivation for reading and reading comprehension. Future research is needed to examine longer term use and its impact on issues such as literacy and comprehension, as well as the impact of access to eBooks on quality of life. Last, the centralized testing location cannot be equated with environments in which people might typically use such devices, including in their homes, on buses, during a workout, and so forth. Ease of use needs to be evaluated in environments that are more ecologically valid. Given these caveats, we believe that the study illustrates the potential benefit to people with intellectual disability of having access to cognitively accessible electronic and information technology.
Wehmeyer et al. (2004) documented that, in too many cases, issues pertaining to the characteristics of people with intellectual disability are ignored in the design of technology. Of particular importance in such designs are features like flexibility, simplicity, and tolerance for error that enable cognitive access. For example, devices like the one used in this study ensure flexibility in use by providing options that accommodate for users' accuracy and precision and adapt to a user's pace. Eliminating the need for a stylus and the concomitant requirement for accuracy and, instead, designing a broader “active button” area that can be selected using one's finger, as was done with the accessible reader, is one example of design features ensuring flexibility and simplicity. Simplicity and intuitive use are important design elements for people with cognitive disabilities. Many devices, particularly software, tend toward greater complexity as a market-driven feature and in so doing limit the device's use by people with cognitive impairments. Cognitively accessible devices and programs address this by allowing the set up of the program to vary according to complexity of the user's needs. As few or as many buttons or icons can be shown as needed; buttons can include both text identification and audio instructions identifying their use; and having buttons fade or disappear after their function has been activated and may not be needed any longer (e.g., direction arrows) all provide a simpler and more intuitively used device. Last, tolerance for error is an important feature. In many devices, a mistake results in the termination of the session or results in the unsuccessful use of the device in some way. So, for example, with the accessible reader hardware, buttons on the Pocket PC are automatically disabled when running to prevent accidental activation of other programs and, consequently, termination of the audiobook reading session.
It is notable that participants both reported and demonstrated their enjoyment with their success in using the system to be able to access audible books. Participants commented on the ease of use of the accessible reader (“It's pretty easy, I like this,” “This is a piece of cake,” “It's easy to use. It reads to you.”) and their general satisfaction with using a palmtop PC with any of the three readers (“I like those little computers” or “I'd like to get one of these for Christmas”). Mainly, though, when asked to comment on the process, what participants referred to was not the device or the software but the fact that it opened the door to literacy for them. Participants observed, “I like the stories and the music,” “I love stories,” “That is a good story,” and “Moby-Dick [Melville, 1851] is my favorite.” One participant observed that he could listen to the books on the bus, because it took him “forever” to get home.
This article is based on work supported by the U.S. Department of Education under Project R305S040023. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views or policies of the Department of Education. The opportunity to interview support professionals was critical to the success of this project. We would like to thank the staff at the Transition Program of Colorado Springs School District 11; MOSAIC, Inc.; and Goodwill Industries, who provided opinions and feedback and otherwise facilitated various tasks in the project. In addition, we would like to thank Audible for supporting this research project by providing the necessary technical information and code to interface with books in Audible's proprietary book format. Last, we would like to thank those individuals who volunteered to participate in the Rocket Reader study and other testing activities conducted as part of this project. Without their help, the project could not have been accomplished.
Authors: Daniel K. Davies, MA (firstname.lastname@example.org) President and Founder; Steven E. Stock, MA, Vice President; and Larry R. King, PhD, AbleLink Technologies, Inc., 618 N. Nevada Ave., Colorado Springs, CO 80903. Michael L. Wehmeyer, PhD, Professor, Department of Special Education, University of Kansas, Lawrence, KS
Daniel K. Davies was the principal investigator on this project, which was funded by the Institute of Education Sciences. Steven E. Stock is a co-researcher on all AbleLink Technologies' research projects. The authors from AbleLink Technologies have used these and other research findings to develop a commercially available e-book software to promote access to electronic books for individuals with cognitive disabilities.