Skip to Content

AFBAmerican Foundation®
for the Blind

Expanding possibilities for people with vision loss

Assessing Writing Software Tools for People with Vision Impairment, Learning Disability, and/or Low Literacy

Rose M. Angelocci
Barbara G. Connors
University of New Orleans Training, Resource, and Assistive-Technology Center

Introduction

Review of the literature reveals that it is difficult to identify and assess people who have a visual impairments and learning disabilities (VI/LD) and to provide appropriate intervention strategies. We evaluated five writing software tools designed for people with learning disabilities to determine their accessibility by and usefulness for people with visual impairment and low literacy. Both strengths and limitations in features and accessibility were identified. Several programs are appropriate for use by people with residual vision. Speech recognition appears to be the most promising accommodation tool for a person who is totally blind. It is postulated that software that can address the needs of people with learning disabilities in addition to visual impairments may help raise their literacy levels. The American Foundation for the Blind Bridging the Gap grant assisted us in this effort.

Definitions

There are generally accepted and easily understood definitions of visual impairment. Visual impairment refers to a range of visual loss that requires adaptations for learning in a variety of environments (Silberman & Sowell, 1998). According to the Individuals with Disabilities Education Act (IDEA) (P.L. 101-476), a visual disability is a visual impairment that, even with correction, adversely affects a child's educational performance. Legal blindness has also been defined as a best corrected central visual acuity of 20/200 in the better eye, or a visual field of 20 degrees or less.

However, there have always been controversies in the delineation and identification of learning disabilities (Scruggs & Mastropieri, 2002). Although learning disabilities have been defined in numerous ways, the major constructs of the definition of specific learning disorders in the IDEA Amendments of 1997 (P.L. 105-17) include the following:

  • There is a disorder in one or more of the basic psychological processes, such as memory, auditory perception, visual perception, oral language, and thinking.
  • There is difficulty in learning, specifically in speaking, listening, writing, reading, and mathematics.
  • The problem is not due to other causes, such as vision or hearing impairment; motor handicaps; mental retardation; emotional disturbance; economic, environmental, or cultural disadvantage.
  • There is a severe discrepancy between apparent potential for learning and a low level of achievement. (Identification criteria of learning disability in most states are determined according to a discrepancy model, in which there is a difference between achievement and measured intellectual potential of between one and two standard deviations.)

An individual who has a visual impairment and a learning disability (VI/LD) is defined by Silberman and Sowell (1998) as having sensory loss, as well as significant difficulties in listening, speaking, reading, writing, reasoning, or calculating. Because some visual disabilities have a neurological origin, and also because learning disabilities and visual disabilities in children who were premature babies occur at a higher rate than is normally the case, there are physiological reasons to expect a relatively high rate of co-occurrence of VI and LD. Existing studies indicate that between 14 percent and 65 percent of students with visual disabilities also have learning disabilities (Erin & Koenig, 1997). This is a higher percentage than the estimated 2 percent to 10 percent occurrence of learning disability in the general population.

Identification and Characteristics of VI/LD

Erin and Koenig (1997) state that a visual impairment seems easier to define, comprehend, and treat than a learning disability. Therefore, a visual impairment may be the only disability diagnosed, when both conditions are present or a learning disability only. They further suggest that learning disabilities may be overlooked in students with low vision and low literacy skills for several reasons:

  • Learning difficulties may be attributed to visual functioning.
  • Visual impairment is detected earlier in life and is more easily observed.
  • Atypical learning patterns can easily be overlooked in students with VI.
  • Low vision may be a more acceptable diagnosis to some people than a learning disability.

The classroom teacher may be the first person to observe a visually impaired student's academic difficulties and request evaluation. Identification of people with VI/LD is problematic because of the traditional use of the discrepancy method of diagnosis. In addition, standardized instruments are typically developed without inclusion of visually impaired people in the normative sample. And application of non-standardized procedures, or modification of standardized instruments in new ways, raises issues of validity (Layton & Lock, 2001).

Identification of VI/LD is also problematic because a person with a visual impairment and a person with a learning disability may appear to have similar academic difficulties. It is even more difficult to identify VI/LD in a student who is blind. Erin and Koenig (1997) indicate, for example, that a student with blindness who has auditory processing difficulties, or one who cannot blend and pronounce words read in braille after easily identifying the letters, might be suspected of also having a learning disability. Layton and Lock (2001) recommend the use of a mixed methodology assessment including both quantitative and qualitative data. Characteristics of VI/LD students may include:

  • Academic achievement below expected performance.
  • High distractibility and lack of attention to task.
  • Avoidance behaviors, such as headaches, talking, or passivity.
  • Problems in learning related memory, perception, organization, concrete thinking, perseveration, and/or language (Erin & Koenig, 1997).

Intervention Strategies for VI/LD

Intervention strategies or approaches recommended by Erin and Koenig (1997) consist of vision enhancement, environmental adaptations, learning strategies, learning opportunities, and compensatory skills. Compensatory skills and accommodations generally used by people with VI/LD consist of audiotapes, human readers, reading machines, talking calculators, talking dictionaries, video magnifiers or closed-circuit television systems (CCTV), and computers with screen reading and magnification capabilities. Writing tools or strategies typically used by people with visual impairments include dark-line paper, raised-line paper, felt pens, braille, dictation, and word processing with assistive technology.

Writing software tools typically used by people with learning disabilities consist of talking word processing programs, word prediction programs, organizational programs, and speech recognition programs. Speech synthesis has been demonstrated to enable students to identify a high percentage of errors in capitalization, spelling, word usage, and typographical errors (Raskind and Higgins, 1995). Speech recognition programs have been shown to improve writing by allowing many students with LD to use more developed oral vocabularies and to experience freedom from concern for spelling, because words are automatically spelled correctly (Higgins and Raskind, 1995). Raskind and Higgins (1999) also suggest that programs designed by assistive technologists appear to serve remedial functions in word recognition, spelling, and reading comprehension, as well.

Method

This project focused on the use of writing software tools traditionally used by people with LD in light of their application for people with visual impairments and learning disabilities or low literacy skills. Five writing software tools typically used by people with learning disabilities only were reviewed: Co:Writer 4000, Write:Outloud Version 3, Draft:Builder, WYNN Wizard 3.1, and Naturally Speaking Professional. Two programs designed to make written matter accessible to those with visual impairments were evaluated for their compatibility with each writing tool: JAWS for Windows 4.5 screen reading software and ZoomText Xtra 7.1 screen magnification software. The JawBone program provided a bridge between JAWS for Windows and Naturally Speaking.

The authors evaluated the software packages based on the following criteria. Some users benefit from hearing characters, words, or sentences as they are typed. It also may be important for one to hear what is being erased when using either the Delete or Backspace keys. When moving through a document, a user may wish to review the file by hearing the previous or next character, word, line, sentence, or paragraph. To remain oriented within a document, the user might want to hear the current character, word, line, sentence, or paragraph and may even wish to have the current word spelled. Menus, dialog boxes, and listed items may be read when encountered. Voice settings such as rate, tone, pitch, volume, or person may be adjusted to a user's preferences. Supplemental sounds can also serve as a cue to alert the user of a special situation such as a misspelled word.

The ability to customize the appearance of a program is also important to both people with visual and/or learning disabilities. Some users can benefit from specific combinations of background and foreground colors. At times, the cursor and mouse pointer may be difficult to locate on screen, and the ability to change their appearance by color, shape, or size can be beneficial. Highlighting the current word with contrasting color to the background and foreground can also assist one in remaining oriented and in focusing one's attention to the current word. The size of text and/or menu items can sometimes be made larger for those who find bigger to be better. Additional space between characters, words, and sentences may also allow information on screen to be more easily read and identified.

Results

Co:Writer 4000

Co:Writer is a talking word prediction program that can be used in combination with a word processing program. Word prediction allows the user to start to type a word and be presented with the computer's "best guesses" about what the user is trying to type. The user can then choose a word from the list or continue typing to further refine the choice list. For Co:Writer, words are predicted based on grammar and phonetic spelling (FlexSpell). The user can also customize predictions by creating specific topic word lists. Co:Writer offers many features to a variety of people with disabilities, including the ability to be a communication aid for those with hearing or speech difficulties, supplemental auditory and visual enhancements for those with visual and/or learning disabilities, and switch access for those with physical disabilities who need an alternative to the keyboard. Co:Writer is available through Don Johnston, Inc. at a cost of $325.

Co:Writer allows the user to change background and foreground colors. The documentation provided with the software even suggests color combinations for different types of visual and learning disabilities. The shape of the cursor could either appear as a solid or open triangle, but the color and size could not be altered. The appearance of the mouse pointer could not be altered through the program. Text size could be adjusted up to "very large," which provided a font size of 48-point type. The size of menu options could not be changed. Co:Writer did not allow spacing between characters, words, or lines to be adjusted.

Co:Writer can be operated by both mouse and keyboard commands. Most keyboard commands were fairly standard, however, a few additional nonstandard keystrokes were available. JAWS for Windows did not seem to enhance the accessibility of the Co:Writer program. JAWS was able to read menus and dialog boxes, but was unable to read words, lines, and backspaced text. JAWS did not read the word prediction list automatically and needed to be supplemented with Co:Writer's speech capabilities.

Co:Writer and ZoomText Xtra Level 1 (magnification only) appeared to work together with little conflict. One difficulty the authors encountered was the inability to scroll down the word prediction list. Enlarging the screen image with a screen magnification program can cause some users to become disoriented on screen because both Co:Writer and a connected word processing program share the screen.

Write:Outloud Version 3

Write:Outloud is a talking word processing program. Writing tools include a spell checker and access to a dictionary with definitions. Auditory and visual feedback alert the user to spelling errors. Graphics can be inserted in the text to support writing activities. Write:Outloud can work in conjunction with Co:Writer or as a stand-alone program. It is available through Don Johnston, Inc. at a cost of $99.

Write:Outloud allows the user to hear characters, words, sentences, and paragraphs as they are typed. It does not allow a document to be reviewed by characters, words, or lines. Write:Outloud provides a keystroke for moving through and reading the document by sentence. Characters are not heard when being erased using either the Backspace or Delete keys. Speech output is not provided in menus and dialog boxes, but toolbar items can be read when hovering the mouse over the item. Write:Outloud provides an auditory cue when a spelling error is made. The current word cannot be spelled aloud within the document but may be spoken along with the suggested replacement when using the spell check feature. The user can adjust some voice settings. There was some difficulty in halting speech instantaneously when desired.

Write:Outloud allows the user to change foreground and background colors. The appearance of the cursor and mouse pointer cannot be adjusted. Increasing font size allows for larger text within a document but not for menu options. Spacing cannot be altered between characters and words but the program does allow lines to be double-spaced or triple-spaced. Words are highlighted when spoken.

Write:Outloud had some conflicts with the JAWS for Windows program. Unless JAWS was loaded first, keystrokes would not be accepted by the Write:Outloud program. Write:Outloud was inconsistent when providing the audio cue to alert the user to spelling errors as they were made. In addition, the spell check feature would not automatically read while JAWS was loaded. JAWS did provide access to menus and dialog boxes.

Write:Outloud can be combined with ZoomText. There was no difficulty in tracking while reading and writing. The ZoomText Xtra Level 2 speech did conflict with the Write:Outloud speech and created "double talk." There are a few keystroke command conflicts between the two programs that can be avoided by using menu options over shortcut commands.

Draft:Builder

Draft:Builder is an organizational writing tool that allows users to create an outline or "map" of their writing with accompanying notes. The program offers writing templates for typical writing assignments. The "bibliographer" helps users create bibliography entries by providing a model entry. Organized data can then be sent to a word processing program. Draft:Builder is available through Don Johnston, Inc. at a cost of $125.

Draft:Builder allows the user to hear characters, words, and sentences as they are typed. It does not provide speech output when reviewing a document or when erasing characters with the Backspace or Delete keys. Draft:Builder will speak for specified items (spelling of words and listed items) only when the "speak" button is pressed. Items in nonstandard dialog boxes can be spoken when passed over with the mouse, but speech is not provided for menus and standard dialog boxes. The user can adjust some voice settings. There was some difficulty in halting speech instantaneously when desired.

The only visual adjustment available to the user is the ability to change text size to small, medium or large. Menus and dialog boxes cannot be adjusted by size. There are no options for changing the appearance of the cursor or mouse pointer. Background and foreground colors cannot be changed. Words are not highlighted when spoken. Spacing cannot be altered.

JAWS for Windows did not provide greater access to the Draft:Builder program because of its design. Draft:Builder is more accessible to mouse users than those using the keyboard. The program contains both standard and nonstandard shortcut keystrokes for performing operations. The program also contains nonstandard dialog boxes that were not automatically read by JAWS. The JAWS cursor would not move when exploring various views within the Draft:Builder program.

Draft:Builder can be combined with ZoomText Xtra Level 1 (magnification only). The ZoomText Xtra Level 2 speech did conflict with the Draft:Builder speech and created "double talk."

WYNN Wizard 3.1

WYNN Wizard (What You Need Now) is a bimodal text reading program with scanning and editing capabilities. Writing tools include spell checker, a dictionary with definitions, word prediction, and an outline feature. Built-in study tools allow the user to highlight portions of text, place bookmarks to easily find important text, and make marginal notes including a voice recorded notation for a particular statement within the document. WYNN allows the user to access the internet and send and receive e-mail messages. It is available through Freedom Scientific at a cost of $995.

WYNN allows the user to hear characters, words, and sentences as they are typed, but does not allow the user to hear what is erased when the Backspace or Delete keys are used. WYNN announces the previous or next word, line, sentence, or paragraph when reading a document (Read Only mode) but not when editing a document (Edit mode). The current word can be spelled aloud. Menu items are spoken, but "grayed items" and associated shortcut keys are not identified. Dialog boxes are spoken, but not all items are identified within a box, and descriptors (such as the "button" label) are not given to identify each item spoken. Toolbar items can be announced when the mouse is hovered over them. Listed items, such as in the word prediction feature, are announced, but the associated function key needed to choose the item is not announced. When typing in the outline mode, the outline level is not announced. WYNN does allow the speech to be adjusted by rate and choice of male or female voice.

WYNN allows the user to change background and foreground colors. No options are provided for changing the appearance of the mouse pointer and cursor. The text size within a document can be set anywhere from 8 to 72 point type. The size of the text in the menus cannot be changed. The user can customize either words, lines, or sentences to be highlighted as they are spoken when reviewing a document in Read Only mode, not when editing. The user can have text displayed with additional spacing between characters, words, or lines as desired. To focus attention the user can choose to highlight portions of text in a specific color or mask surrounding text.

With JAWS for Windows added, characters, words, lines, sentences, and paragraphs could be announced while editing. In addition, the erased character could be spoken when using the Backspace or Delete keys. In some instances, the JAWS voice would conflict with the WYNN voice and "double talk." There were also some keyboard command conflicts between the two programs that can be avoided by using the JAWS "by-pass" feature or executing commands using menu options. The word prediction list did not speak automatically and had to be reviewed using the JAWS cursor. There was a time difference between when words were highlighted and when JAWS announced them.

With ZoomText loaded, the screen automatically scrolled while in Edit mode but was not able to follow the highlighted text within the Read Only mode. The ZoomText voice conflicted with the WYNN voice and caused "double talk." Some keyboard command conflicts arise between the two programs that can be avoided by redefining keystrokes or executing commands using menu options.

Naturally Speaking

Naturally Speaking is a voice recognition program that allows the user to use continuous or natural speech patterns to enter data and execute commands. The user can achieve dictation speeds of up to 160 words per minute. It is compatible with many Microsoft Windows' applications. Naturally Speaking Professional is available from ScanSoft at a cost of $695.

Naturally Speaking has a built-in text-to-speech (TTS) option that allows words, lines, sentences, and paragraphs that have been dictated to be spoken. Speech is not provided in menus and dialog boxes. Naturally Speaking can be combined with JAWS for Windows to provide greater auditory access. JawBone is a program that acts as a bridge between Naturally Speaking and JAWS for Windows. It is available from Next Generation Technologies at a cost of $590. JawBone provides access to the JAWS reading and navigation commands and allows JAWS to echo dictation. JawBone provides access to the Naturally Speaking training dialog box, correction dialog box, and other utilities. Audio cues in the form of wave files play when the microphone is turned on and off.

Naturally Speaking can also be accessed by ZoomText users. If Naturally Speaking is detected at installation, ZoomText will modify the command set for Naturally Speaking to include spoken commands for ZoomText commands for magnification, view, navigation, speech, and the doc reader. ZoomText tracks the cursor while the user is dictating. Scrolling is required for reviewing the document and making corrections. Scrolling can be initiated with verbal commands but was found to be difficult to stop with precision. This may prove to be disorienting for some users.

Discussion

The goal of this project was to evaluate some of the writing software tools used by people with learning disabilities to determine the applicability of these tools to people with visual impairments, learning disabilities and low literacy skills.

Co:Writer, Write:Outloud, Draft:Builder, and WYNN may be beneficial to some people with low vision who are assisted by supplemental speech. Although each program still has limitations for low vision users, some people may be able to use them with built-in visual features and familiarity with the screen layout. All programs evaluated seemed to work with ZoomText Level I (magnification only). When used with ZoomText, the programs were limited by the same factors that are inherent in all screen magnification-what is displayed on the screen is restricted and the low vision user needs to scroll more frequently than an individual with fully functional vision using no magnification. Naturally Speaking would be more effective for low vision users who are not limited in the use of their hands, as scrolling cannot be stopped instantaneously with a verbal command. The combination of using one's voice to dictate and using the keyboard or mouse to review would make the process of writing more effective for a ZoomText user.

Co:Writer, Write:Outloud, and Draft:Builder do not have sufficient built-in speech to be effectively used by a person who is totally blind. JAWS for Windows did not make these products more accessible. WYNN offered more accessibility for those who rely on speech, but still had several limitations that diminish its effectiveness as a writing tool for someone who is blind. JAWS did increase access to the WYNN program and could be managed by a skilled JAWS user if desired. Naturally Speaking seems to be the most promising writing tool reviewed here that can be accommodated for a person who is totally blind. Again, this would require great skill in that several software packages must be layered to create an effective tool.

Further research is needed to find literacy solutions for people with visual impairments and learning disabilities or low literacy skills. Other software packages, such as Kurzweil 3000 and TextHelp, could be evaluated by the same criteria herein applied. The use of an electronic braille display could be tested as another mode by which these programs may be accessed for people with visual impairments. Training in these writing software packages with appropriate accommodations should be provided to determine user and training concerns. Advantages offered by these software packages over standard word processing tools can be discovered through wider use or adapted applications. Finally, literacy outcomes need to be evaluated to determine if these products have a similar impact on literacy levels for people who are visually impaired as they appear to have with individuals who are sighted and learning disabled.

Product Information

Co:Writer, Write:Outloud, Draft:Builder
Don Johnston, Inc., 26799 West Commerce Dr., Volo, IL 60073 (800-999-4660)
www.donjohnston.com

WYNN (What You Need Now), JAWS
Freedom Scientific, 11800 31 Court North, St. Petersburg FL 33716 (800-444-4443)
www.freedomscientific.com/wynn

Dragon Naturally Speaking
ScanSoft Inc., 9 Centennial Dr., Peabody, MA 01960 (978-977-2000)
www.scansoft.com

ZoomText
Ai Squared, P.O. Box 669, Manchester Center, VT, 05225 (802-362-3612)
www.aisquared.com

JawBone
Next Generation Technologies, Inc. (425-744-1100)
www.ngtvoice.com

References

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.) Text Revision. Washington, DC: Author.

Harley, R., Truan, M., & Sanford, L. (1987). Communication skills for visually impaired learners. Springfield, IL: Charles C. Thomas.

Higgins, E., & Raskind, M. (1995). Compensatory effectiveness of speech recognition on the written composition performance of postsecondary students with learning disabilities. Learning Disability Quarterly, 18, 159-184.

Individuals with Disabilities Education Act Amendments of 1997, P.L. 105-17, 105th Congress.

Layton, C., & Lock, R. (2001). Determining learning disabilities in students with low vision. Journal of Visual Impairment and Blindness, 95, 288-299.

Raskind, M., & Higgins, E. (1995). Effects of speech synthesis on the proofreading efficiency of postsecondary students with learning disabilities. Learning Disability Quarterly, 18, 141-158.

Raskind, M., & Higgins, E. (1999). Speaking to read: The effects of speech recognition technology on the reading and spelling performance of children with learning disabilities. Annals of Dyslexia, 49, 251-281.

Scruggs, T. & Mastropieri, M. (2002). On babies and bathwater: Addressing problems of identification of learning disabilities. Learning Disability Quarterly, 25, 155-168.

Silberman, R. & Sowell, V. (1998). Educating students who have visual impairments with learning disabilities. In S. Sacks & R. Silberman (Eds.), Educating students who have visual impairments with other disabilities (pp. 161-185). Baltimore, MD: Paul H. Brookes.

services icon Directory of Services

book icon Featured Book

Braille LiteracyA Functional Approach

Braille Literacy:

Join Our Mission

Help us expand our resources for people with vision loss.