This paper was posted on the LD Online website, http://www.ldonline.org, as part of the October 1998 Focus on Technology. Permission to reprint is granted to the Georgia Advocacy Office for noncommercial use by the author.

Reading Software for Students with Learning Disabilities:
Hypermedia-Based Children's Literature
Rena B. Lewis, Ph.D.
Project Director, Project LITT
San Diego State University

The fourth grade boy sits at the computer ready to read his book. It is a CD-ROM program, a "talking storybook." He starts the program and turns to the first page of the story. The screen comes alive with color, sound, animation, and speech. The characters in the illustrations move and speak as the narrator reads the text aloud. Then, it's the boy's turn to interact with the text and graphics. "Neat," he says.

Hypermedia-based children's literature programs, sometimes called "talking storybooks," are one of the most popular types of software today. These programs read stories aloud to students in realistic digitized speech, colorful graphics accompany the text, and students often can interact with both text and graphics. Examples are the programs in series such as Living Books, Disney's Animated Storybooks, Reader Rabbit's Reading Development Library, Discis Books, Magic Tales, and WiggleWorks.

Project LITT, Literacy Instruction Through Technology, is a three-year research project funded by the U.S. Office of Special Education. The overall purpose of Project LITT is investigation of the effectiveness of hypermedia-based children's literature software in improving the reading skills of students with learning disabilities. This article describes the project, the body of software under study, and preliminary results from several empirical studies. You may choose to read the article from start to finish or jump to one of these topics:

The Potential of Hypermedia-Based Children's Literature Software
Hypermedia-based children's literature has several potential advantages for students with learning disabilities who are struggling to acquire basic reading skills. First is the motivational appeal of this body of software. With its dazzling graphics, realistic sound, and plentiful opportunities for interactions between the learner and the task, this type of software has the capability to capture and hold students' attention. As Erickson and Staples (1995) reported, even students with autism respond to the attractiveness of these programs with increased attention to the reading task. This level of motivational value may increase the probability that reluctant readers will persevere in their interactions with text. This would be a particularly valuable outcome because repeated readings of the same text have been found to be of value for students with learning disabilities (Sindelar, 1987).

Also, hypermedia-based children's literature offers students text that is speech-enhanced and therefore cooperative (Boone, Higgins, Falba, & Langley, 1993). The speech quality is high and, depending on the program, students may have access to either whole-word feedback or both whole-word and segmented feedback. As Wise and Olson (1994) pointed out, whole-word feedback appears to provide the most support to students with severe reading problems. Segmented feedback also appears valuable because of its positive effects upon students' phonological decoding skills.

In addition, this software is a computer translation of children's literature. When transformed into computer-mediated "books," the quality of the texts and illustrations are preserved. Texts are typically heavily illustrated; also, they are often predictable and include narrative features such as repeated lines and rhymes. These features, like software speech enhancements, increase the cooperativeness of the text. Comprehension is aided because of the graphical cues and the predictability of the text. Also, if a computer-mediated book is used as a springboard for instruction in skills such as decoding, that instruction is easily "anchored," as Hasselbring and his colleagues (e.g., Bottge & Hasselbring, 1993) use the term, to the student's experiences with that piece of children's literature.

Moreover, hypermedia-based children's literature programs offer new opportunities to students with learning disabilities whose first language is Spanish, rather than English. Many programs allow students to hear the story read aloud in either language; some, like the Discis Books series, provide Spanish assistance in English word-reading tasks. Researchers such as Ruiz (1989, 1995) suggest that the communicative competence of bilingual exceptional students can be enhanced when the classroom context emphasizes communication, centers on students' background knowledge and experiences, allows for student initiations and student-directed discourse, and involves whole texts, rather than fragments. Hypermedia-based children's literature in English and Spanish could certainly contribute to establishing a more optimal learning environment for these students.

Finally, the vast majority of this software is general education software. Because it is widely used in typical classrooms, it reflects the reading experiences of typical children. Use of this software as a means to ameliorate the reading problems of students with learning disabilities would allow them to participate in the same types of reading activities as their peers. And, when students with learning disabilities are educated in inclusionary settings, instruction with talking storybooks would call less attention to their reading difficulties than similar efforts with "special" software. Armstrong, Brand, Glass, and Regan (1995) recognized these advantages when they included hypermedia-based children's literature among their recommended general education software titles for students with special needs.

It is also important to recognize that there are several potential disadvantages to hypermedia-based children's literature software for students with learning disabilities. First, these programs rely on discovery learning. Because they are hypermedia, the programs do not lead the student through a carefully sequenced series of instructional activities. Students choose their own paths through the programs, and their interactions with text elements may be quite limited. Unfortunately, mere exposure to text is unlikely to increase the reading proficiency of students with learning disabilities. As observed by Lewis and Doorlag (1999), these students "will likely require explicit instruction in skills such as decoding" (p. 246). Thus, it is likely that students with learning disabilities will require instructional support in order to maximize the benefits they might derive from this type of software.

Second, the very elements of hypermedia-based children's literature that make it appealing and motivating to young readers may decrease its usefulness for students with learning disabilities. There is wide variation in the degree of emphasis this software places on the reading task. In some programs, students can virtually ignore the text and concentrate on interacting with its illustrations. However, as Lewis (1998) points out, "10 seconds of reading followed by 10 minutes of play is not a good use of instructional time" (p. 22). In addition, some students with learning disabilities are characterized by their distractibility (Reid, Maag, & Vasa, 1994); poor skills in marshaling attentional resources will not assist students as they attempt to navigate through programs filled with highly entertaining distractions. These limitations underscore the need for instructional support in the use of this type of software.

Third, students with learning disabilities may encounter difficulty in the use of speech-enhanced text to support the reading process. As MacArthur and Haynes (1995) and Wise and Olson (1994) observed, students with learning disabilities do not consistently take advantage of the support provided by speech-enhanced text. In Olofsson's (1992) study (as reported by Lundberg, 1995), students with disabilities used speech supports less often than typical readers, despite the fact that they were less skilled in the decoding process. Again, instructional support, above and beyond the cooperative text provided by the software, is required to optimize the benefits that students with learning disabilities may obtain from hypermedia-based children's literature.

Fourth, because this software is designed for general education audiences, it is geared to the interest levels of students achieving at grade level. Students with learning disabilities, however, typically read below grade level. To match their interest levels with their reading skills, high-interest low-vocabulary reading materials are required. Unfortunately, there is a scarcity of high-interest low-vocabulary reading software in the talking storybook format.

In summary, hypermedia-based children's literature has many potential advantages for improving the reading performance of students with learning disabilities. It is appealing general education software that presents high quality children's books in an interactive format. In addition, it includes a variety of features that support the reading process including graphical cues, predictable text, and speech enhancements in both English and Spanish.

However, there are potential drawbacks and many of these center around the interactive nature of this type of software. Because it is hypermedia-based, the student directs the flow of interactions with the text and other elements of the program. However, interactive graphics may compete with text for students' attention. And, when students do attend to the text, they may fail to take full advantage of the speech enhancements available to support them in the reading process. Students with learning disabilities may lack the strategies needed to successfully negotiate hypermedia-based children's literature software in order to improve their reading skills.

At present, it is unclear whether the potential advantages of hypermedia-based children's literature as an instructional tool for students with learning disabilities outweigh its possible disadvantages. Final judgments must be postponed until research evidence becomes available. Project LITT's purpose is to gather the data needed to determine the instructional usefulness of talking storybooks for students with learning disabilities.

The Software Search: What's Available?
In order to study the effectiveness of hypermedia-based children's literature, Project LITT began by conducting a nationwide software search. The goal was to identify and evaluate all commercially available talking storybook programs. As of October 1998, we have located more than 300 titles. (Software Profiles for each of these programs are available on the Project LITT website, described later in this article.) Project staff reviewed each program, story texts were evaluated for readability, and students with learning disabilities and their teachers critiqued a sample of representative programs. Programs were included in the collection if they were CD-ROM based, they included text, and that text could be read aloud. Excluded were programs with nonfiction text (e.g., science, social studies) and those with estimated readability levels higher than grade 8.

The first major conclusion that can be drawn from the software search is that there is a large number of talking storybook programs currently available. In planning the project, we had estimated that 50 to 75 such programs existed; our estimates were clearly too low. In 1996-97, more than 250 programs were identified; in the next year, approximately 50 additional programs were located. However, the rate at which new programs appear on the market seems to be slowing; in the summer of 1998, for example, fewer than 5 new programs were published.

The second major conclusion from the software search is that this body of software is extremely heterogeneous. Talking storybooks vary in several ways including their length, difficulty level, and the types and numbers of opportunities they provide to students to interact with text, graphics, and activities. The paragraphs that follow describe the body of software in an attempt to point out the areas in which variability occurs along with some of the commonalities among programs.

Titles. All types of stories are available in CD-ROM format. Some talking storybooks are software adaptations of award-winning children's books, others are traditional children's stories and fairy tales, and some are new texts especially written for the software format. Several duplicate titles were located in the software search, and the most common were traditional tales such as Jack and the Beanstalk and Goldilocks and the Three Bears.

Computer platform and price. Most talking storybook CD-ROMs (72%) are dual platform, i.e., available for both Windows and Macintosh computers. Most are relatively inexpensive (less than $50) (79%). Prices do vary, however, from programs costing less than $20 to one multi-CD-ROM instructional package costing almost $2,000.

Intended market. Talking storybooks are produced for both home and school markets. Home programs tend to be less expensive, more likely to include many opportunities for interactions with graphics, and more likely to contain game activities. School programs, in contrast, typically offer instructionally relevant features such as glossaries, word processing activities, and reading skill lessons.

Length of stories. There is enormous variability in the length of on-screen stories. Some stories are very brief, including fewer than 50 words. Others contain texts of more than 12,000 words.

Availability of print books. About one-fourth of programs (27%) include a print book along with the CD-ROM. However, there are often differences in the text and illustrations between the print and software versions. In many cases, the on-screen story is a shortened version of the original book. For example, the Living Books CD-ROM Stellaluna contains 343 words whereas the print book contains 1,201.

Estimated readability level. The readability level of the on-screen text may be different from the print book, and it may also be inconsistent with the age or grade levels recommended by the publisher. We used the Grammatik program to compute estimated Flesch-Kincaid readability levels for all stories in the Project LITT collection. Estimated reading levels ranged from grade 1 through grade 8. On-screen stories were fairly equally distributed between the grade 1-2 range (28%), the grade 3-4 range (39%), and the grade 5-6 range (27%).

Age-appropriateness. The age-appropriateness of programs is related not only to readability but also to factors such as story content and the program's appearance. Stories that are likely to interest and appeal to younger elementary grade students may be written at higher grade levels. For example, Winnie the Pooh and the Honey Tree, one of the programs in Disney's Animated Storybook series, is recommended by the publisher for ages 3 to 8; its estimated readability level is grade 7. Also, some programs incorporate music, cartoon characters, and children's voices that make them most appropriate for young students, even though the story content and reading levels may make more sense for older students.

Availability of languages other than English. Some talking storybooks are bilingual or multilingual (23%). When languages other than English are available, Spanish is the most common although some books provide French, German, and Japanese versions. In some multilingual programs, the story appears in its entirety in each language addressed. For example, there may be a version of the story in English and a separate version in Spanish. However, some programs provide only limited support in a language other than English. For example, students may be able to click on a word in the English text to hear (but not see) its Spanish translation.

Text interactivity. The programs in the Project LITT collection allow students to interact with the text and hear that text read aloud (programs without that feature were excluded). In all programs, students can hear the entire on-screen "page" read aloud. In more than half of the programs (56%), students can also hear individual words read aloud, and these programs were rated "high" in text interactivity.

Interactivity with graphics. Much more variability among programs was seen in the amount of graphics interactivity. The range extended from programs with no opportunities for interactions with graphics to those with literally dozens of opportunities ("hot spots") on each "page." Thirty-eight percent of the talking storybooks offered no opportunities for interactions with graphics, and these tended to be programs designed for school markets (e.g., WiggleWorks, Multimedia Literature). In 17% of programs, there were more than 15 "hot spots" per page, on average; these programs were rated "high" in graphics interactivity. In most cases, students clicked on hot spots to see animations of the graphics and hear speech, music, and/or sound effects. In some program series (e.g., Discis Books, Top Hat Tales), clicking on a hot spot produces a picture label. For example, if the student clicks on the picture of a duck within the story illustration, a label with the word "duck" appears and the word is pronounced.

Features and activities. Many talking storybook programs offer features and activities that support the development of literacy skills. The most typical literacy activity is a word processing feature, provided by 55% of the programs. Reading mini-lessons or activities were found in 38% of programs and word definitions and/or glossaries in 38%. Programs designed for the home market were more likely to contain games and activities such as matching, concentration, mazes, and the like. For example, Hunchback of Notre Dame, one of the titles in Disney's Animated Storybook series, contains five activities including arcade-type games such as "Gargoyles vs. Soldiers" and "Climbing Down the Walls."

Teachers' and Students' Views of Talking Storybook Software
Students with learning disabilities and their teachers were asked to view representative talking storybook programs and share their views on the usefulness of these programs for improving reading skills. Focus group discussions were held with 42 special education teachers and 13 students with learning disabilities.

Students' Views
The students with learning disabilities who participated in the focus groups were in grades 2 through 7. Students were very enthusiastic about the talking storybook programs they viewed. All liked the programs and, when asked what they liked best, students identified features related to the interactivity of the software. For example, several students commented on the animated graphics that brought story illustrations to life. One sixth grade girl said, "It looked like it was really happening." A fourth grade boy observed, "It moves instead of a book that just stays." Students also talked about text interactivity. As one fourth grader said, "I can find a word and it says it." In most cases, students were not able to identify aspects of the programs that they did not like. However, two second graders who viewed a program with no graphics interactivity complained that the pictures didn't move. Students were asked whether they thought that talking storybook programs would help them to read better. Most students felt these programs would improve their reading skills because the story is read aloud, "it shows the words and you can read along with it," and "you can click on a word and it will say it."

Teachers' Views
Teachers had generally positive reactions to the talking storybook programs they viewed during the focus group discussions. However, they were critical of program features that they felt detracted from instructional goals, and they had strong opinions about which features software developers should include when designing programs to enhance students' reading skills. In addition to pointing out that clear directions are essential for student success, teachers' discussions centered around three main areas: the text presented in the stories, the graphics components of the software, and program features related to instruction and individualization.

Text considerations. Many of the programs evaluated by the teachers were based on print books and, in some cases, the books were included with the software package. Teachers felt this was a good practice because it gave students the opportunity to interact with both the traditional print and software versions of the story. Teachers preferred programs that did not deviate substantially from the text of the book. Unfortunately, many children's books are quite long and the text is condensed when the software version is created.

Teachers viewed programs very positively if they offered versions of the story in languages other than English. Spanish is the language other than English that is most often available in talking storybook programs, and the teachers felt this choice was the best for the population of students they served. Programs offering only English text were considered much less desirable than multilingual programs.

Several factors related to the appearance of the text on the screen were considered important by teachers. They preferred text styles and fonts that were familiar to children and easy to read. They believed text should be large enough for it to be seen easily, and there should not be too much text on any one screen "page." The color of the text should be chosen so that the words stand out from the background, particularly when the background is a "busy" graphical illustration. Even more preferable is text on a plain white or neutral background. Several teachers commented on the placement of the text on the screen. In most cases, teachers wanted the text placed in a logical location (e.g., at the bottom of the screen) and preferred that text remain in the same location from screen to screen. Teachers particularly liked programs where the on-screen displays resembled books (e.g., the Reader Rabbit's Reading Development Library Series).

Teachers preferred programs where the text is highlighted as it is read aloud. In general, teachers wanted text highlighted either word by word (to direct learners' attention to individual text elements) or phrase by phrase (to encourage more fluent reading and better comprehension). Less desirable were programs that highlighted entire sentences or each line of text. Least liked were programs where text was not highlighted because teachers believed that students with learning disabilities would not be able to keep their place in the text. The method used to highlight text was also important. Teachers criticized programs where colored highlighting made text more difficult to see.

In all of the programs that teachers reviewed, the text on the screen was read aloud. Teachers considered this feature essential for instruction but voiced some concerns. Although teachers wanted the programs to move along at a brisk pace to keep students' interest, one common complaint was that the narrator read too quickly. Teachers feared that students would have difficulty keeping up and would become lost. Teachers preferred narrators who read with expression and criticized those with monotone or "boring" voices. Also, teachers felt that juvenile or "babyish" voices would be unattractive to older students.

Most programs viewed by the teachers allowed students to hear the text read aloud again (after the narrator had read the page in its entirety). Teachers liked this capability, particularly when students were able to select individual words rather than sentences, paragraphs, or the entire selection on the page. Also highly rated were programs that provided information about word meanings. Included in the software sample were examples of programs that offered definitions and explanation of words in a variety of modes including text, text accompanied by speech, multimedia presentations with brief video clips, drawings or illustrations (for nouns), and animated graphics sequences (for verbs).

Teachers were very sensitive to the content of the story, its value as literature, its appeal to students, and its fit with the goals of the curriculum. Several programs were praised for their inclusion of diverse groups and their multicultural themes. Teachers also carefully examined the structure of the language used in the stories, and those working with less able readers favored predictable and repetitive texts and ones that included rhyming words and high frequency vocabulary.

One major concern was the appropriateness of the programs given the discrepancy between their students' ages and the grade levels at which they were able to read. Teachers criticized stories they perceived as too juvenile in content. They were also concerned that the reading level of some stories might be too high. The stories in the programs reviewed by teachers ranged in readability level from grade 1 to grade 6. However, even some of the stories with the lowest readability levels were considered too difficult for some elementary grade students.

Considerations related to graphics. Teachers evaluated the illustrations accompanying the stories in terms of their appeal to students, age appropriateness, and general quality. Teachers expressed individual preferences for some illustrations; for example, some professionals liked more realistic depictions of the characters in the story rather than cartoon-style drawings. Teachers were particularly concerned that illustrations not be too "primary" and that the age level to which they appealed be congruent with the content of the story and the readability of the text.

The programs that teachers viewed ranged from those with no capacity for interaction with the graphics on the screen to others with very large numbers of interaction opportunities. One way to quantify this is by the numbers of "hot spots" per screen. A hot spot is a screen location which, when selected by a student, produces some type of result. Examples are: (a) the appearance of a picture label (e.g., the word "dog" appears next to a graphic of a dog and the word is read aloud), (b) a short animation sequence involving one of the objects on the screen and sound effects (e.g., a flower sways to music or a dresser drawer opens and bangs shut), and (c) a more prolonged animation sequence in which characters in the story move, speak, and interact accompanied by sound effects and/or music. The programs used in the demonstrations varied from some with no hot spots to those with more than 40 hot spots per screen, on average.

On one hand, teachers were enthusiastic about programs with interactive graphics because they felt they'd be highly motivating to students. Comments included "It makes learning fun," "Grabs kids," and "high interest level." On the other hand, teachers were highly critical of programs that contained "too many" hot spots and those where selecting the hot spots disrupted the continuity of the story. Teachers seemed to prefer programs in which the graphical interaction related to the story (e.g., The Living Books Series), the interactions retold the story (Reader Rabbit's Reading Development Library), or the interactions were simple picture labels rather than animation sequences (e.g., Top Hat Tales, Discis Books). Teachers also liked the WiggleWorks series which offers no graphics interactions. As one teacher remarked, "I like that this focuses on actually teaching reading."

Another concern raised by teachers related to long action sequences that preceded and/or followed the oral reading of the text on the screen. Most felt that, although these sequences related to the story, they distracted students from the reading task.

Other considerations. Three other areas were repetitive themes in teachers' discussions of the programs: games and other activities, instructional features of the programs, and options for individualization.

In general, teachers were not impressed with games and other types of activities embedded within the stories. They felt they disrupted the continuity of the story line and, although entertaining, were too much of a distraction. Most criticized was the Disney's Animated Storybooks series designed for use at home, not school. Teachers' observations included "Too 'gimmicky,'" "more for entertainment than education," and "students would go to game before choosing to read text on their own."

The instruction-related features of the programs typically were adjunct activities which students (or their teachers) could choose to select. Overall, teachers were pleased with these features and felt they contributed to the educational value of the programs. Most common were writing activities such as "blank books" where students could write and illustrate a story. Also available were more structured writing tasks such as the letter writing activity in Reader Rabbit's Reading Development Library (students choose sentence endings) or the "Let Me Write" activity in Arthur's Reading Race from the Living Books series (students modify sentences by selecting objects on the screen).

Teachers also liked programs that provided or collected lists of words to which students could refer. For example, the Discis Books keep a record of each word the student selects. The WiggleWorks programs allow students to build a "My Words" list as they read a story, and those words can then be used in the programs' writing activities.

Other popular features were the ability for students to tape-record their voices as they read stories aloud (e.g., the WiggleWorks series) and activities or options that promote comprehension skills. For example, the Reader Rabbit's Reading Development Library series offers two comprehension features. At the start of each story, the student selects the narrator (e.g., in "The Three Little Pigs," a traditional storyteller, the second pig, or the wolf), and the story then is told from the point of view of that narrator. Also available is a Story Map activity where students put pictures of the events in the story in order.

Art activities were available as part of the writing activities in several programs, but teachers did not regard these as important features. However, teachers were enthusiastic about The Art Lesson, one of two programs in MECC's Stories That Click series. This is an autobiographical story read aloud by Tomie de Paola, the author and illustrator. It includes 14 art activities embedded with the story, and teachers felt these would motivate students, particularly those with interest and talent in art. Teachers also liked the two extension activities in the program, an interview with the author/artist and a tour of his studio, both of which include brief video tape segments.

Teachers liked programs that gave them some control over instructional parameters and enabled them to customize program features to the needs of individual students. Examples of customization options available in some programs are the ability to change the size, style, and color of text; to turn the highlighting on or off, to modify the speed with which text is read aloud; and to determine what portion of the text is read aloud automatically (e.g., the entire selection or individual sentences). Unfortunately, only a few software series offer features such as these, and no program allowed teachers to control the elements they found most intrusive to the learning process: games embedded within the story and the number of hot spots per page. Teachers also expressed a desire for record keeping capabilities within programs to help in monitoring student progress; again, no program offered this feature.

Teachers' recommendations. Teachers recommended that great care be taken in selecting talking storybook programs for use in teaching reading skills to students with learning disabilities. The suggestions that follow summarize their major points:
o Talking storybooks are typically rich and engaging programs that appeal to students, keep their attention, and motive them. However, beware of programs with more entertainment than educational value.
o Select programs that are enhanced versions of excellent storybooks for children. Don't settle for poor or mediocre children's literature because it is found on a CD-ROM disc.
o Give preference to programs where the focus is on the story (rather than on dazzling graphics, superfluous hot spots, or unrelated activities).
o Choose programs that are appropriate for students' ages in content, text, graphics, and narration. Avoid programs where one element (e.g., the content of the story) is clearly discrepant from another element (e.g., the graphics).
o Consider the readability level of the story and other characteristics of the text (e.g., appearance, interactivity).
o Carefully evaluate the graphical components of the program and whether they enhance or diminish the reading experience.
o Whenever possible, select programs that are both age-appropriate and skill-appropriate for students.
o Select programs with useful instructional features such as writing activities; avoid programs where game-like activities interfere with the story.
o Look for programs where teachers can control important instructional parameters such as the size of the text and the speed at which text is read aloud.

Software Profiles: The Project LITT Website
The Project LITT website contains information about the project and Software Profiles for the more than 300 talking storybook programs in the collection. The address of the website is: http://edweb.sdsu.edu/SPED/ProjectLitt/

The Software Profiles are descriptions of talking storybook programs. Each Profile contains information about the program's name and the software series to which it belongs (if applicable); the publisher, price, and platform; the content of the story, its estimated readability level, and actual length; language(s); the degree to which text and graphics are interactive; navigation through the program; and program features such as games, definitions and glossaries, and reading and writing activities.

The sample Software Profile below describes a new program in the Living Books series, Arthur's Computer Adventure. Please note that the underlined text in this example is not linked. However, on the Project LITT website, underlined text is linked so that, for example, it is possible to go directly to the websites of software publishers.

Sample Software Profile
Title: Arthur's Computer Adventure
Series: Living Books
Publishe:r Brøderbund (www.livingbooks.com)
Platform: Macintosh, Windows
Price: inexpensive (less than $50)
Age/Grade Level Rec'd by Publisher: Ages 3 to 7
Print Book: N/A
On-Screen Story: 401 words of interactive text; estimated Grade 3 readability level
Language: English
Summary of the Story: Arthur loves to play the game Deep Dark Sea on his mother's computer. Once he starts playing, he can't stop. He and his friend Buster decide to play the game even though Arthur's mother has told him not to use her computer. The trouble starts when the computer screen goes blank.
Story Interaction Modes: There are 2 interaction modes. In the more passive mode (Read to Me), the learner simply listens as the story is read aloud. In the more active mode (Let Me Play), the learner can interact with the text and graphics in the story after each page has been read aloud.
Text Interactions: The text is highlighted word by word as it is read aloud; each word remains highlighted until the entire sentence is read. In the Let Me Play mode, the page is read aloud automatically. Then learners can ask to hear the entire page or a single word read aloud again. On some screens, text appears and is read aloud but it is not interactive; the program moves immediately to the next screen.
Text Interactivity Rating: HIGH; 12 interactive text screens
Interactions with Graphics: In the Let Me Play mode, learners click on "hot spots" within the graphics. Actions result including animation of the graphics, sound effects, speech, and music. Learners can also access 5 activities from story pages.
Graphics Interactivity Rating: HIGH; 12 interactive graphics screens; 23 "hot spots" per screen, on average
Amount of Animation in Program HIGH
Navigation through the Program: The program begins with a main menu where learners choose the interaction mode. On each story page are arrows to move to the previous or next page. Clicking on the page number brings the learner back to the main menu. The "Pages" choice at the main menu allows learners to go to any page within the story. The "Activities" choice allows learners to go directly to any of the five activities. Learners can also choose "Quit" to exit the program.
Features: o Print Option -- It is possible to print story pages. Press the ALT or Command key to display the menu bar; choose Print from the File menu. o Toy Copter -- Use the remote control to fly the helicopter to the target and drop, then rescue the firefighter. o Deep Dark Sea -- In this computer game, help the diver find the treasure; hazards include running out of oxygen and sea creatures such as sharks and octopuses (3 levels of difficulty). o Frankenfish -- Learn about fish by clicking on their names. Make silly fish by rearranging heads, bodies, and tails; stickers can also be added. Fish pages can be printed. o Treasure Hunt Game -- One or two can play this board game. Players choose to be Arthur or Buster, then select one of three levels; the object is to reach the treasure first. In all levels, players match words and pictures, although difficulty increases. The winner gets to decide how to spend the treasure. o D. W.'s Store -- Visit the store to buy accessories for toys; when learners run out of money, they can earn more at the café by following recipe directions.
Definitions and/or Glossary: no
Writing Activities: no
Reading Activities: yes
Description of the Series: The Living Books series is a large (>10) collection of programs designed for learners in the elementary grades. Books are primarily by contemporary authors such as Marc Brown, Mercer Mayer, and Stan and Jan Berenstain. Most programs include a print book along with a CD-ROM (although the on-screen story is usually shorter than the book version). Each CD-ROM contains one story. All programs are in English and several offer versions of the story in other languages, with Spanish being the most common. (Search for other titles in the series.)

Software Profiles can be accessed at the Project LITT website in two ways. The first method is to find a program in one of the five Software Lists:
o Titles of talking storybooks -- An alphabetical list of the titles of individual stories.
o Names of software series -- A listing of storybook titles by the software series in which the titles are found. Most, but not all, programs are part of a series.
o Readability estimates -- A listing of storybook titles by estimated readability levels.
o Software features -- A listing of storybooks by types of instructional features (e.g., glossaries, writing activities, reading skill lessons).
o Languages other than English -- A listing by language of storybooks with versions in languages other than English.

It is also possible to search the data base for programs meeting specific needs. The search parameters are language, estimated readability, text interactivity rating, and graphics interactivity rating. For example, a teacher or parent could search for programs in all languages with an estimated reading level of grade 4, high text interactivity, and low graphics interactivity.

The Project LITT website will be updated throughout the project as new programs become available. We welcome your input on the design of the website, the Software Profiles, and any programs that we've overlooked.

Observational Studies of Students with Learning Disabilities
The second phase of Project LITT focused on the ways in which students with learning disabilities interacted with hypermedia-based children's literature programs and the gains in reading skills that resulted from these interactions. In the first study, students were observed as they engaged in unstructured interactions with the software. In the second study, increasing levels of instructional support were introduced to determine if such support could result in improved reading performance.

Students' Unstructured Interactions with Talking Storybooks
Six elementary grade students with learning disabilities were introduced to a range of hypermedia-based children's literature programs, then observed as they interacted with these programs under unstructured classroom conditions. Programs were carefully chosen to represent the heterogeneity of this body of software. Included were:
o Type 1: 6 programs with high text interactivity, no embedded games or activities, and varying interactivity with graphics
o Type 2: 6 programs with varying text interactivity and games and activities embedded within the story
o Type 3: 6 programs with varying text interactivity, definitions or glossary features, and picture label "hot spots"

Each student interacted with three different programs of one type, each for a total of 2 hours over 4 days. Students were observed as they worked at the computer and videotaped. Before each intervention began, students' reading skills were assessed by asking them to read aloud 50 words selected from the story in the software program. At the end of each intervention, students read the words again, retold the story in their own words, and answered questions about their perceptions of the software.

The intervention in this study included a range of software programs because we hypothesized that some types of programs contained greater numbers of "attractive nuisances" than other types. When students with learning disabilities interacted with programs with many opportunities for interactions with graphics and those with game-like activities, we believed students would be more likely to be distracted from the reading task and thus less likely to learn new words. This did not prove to be the case.

Preliminary analyses of the data from this study suggest two important findings:
o Students with learning disabilities chose nonreading activities over reading activities when interacting with all types of talking storybook programs.
o In unstructured interactions with talking storybook programs, students with learning disabilities did not show appreciable gains in reading skills.

Analyses of videotapes indicate that students spent at least 40% of each 2-hour instructional period engaged in nonreading activities such as interacting with hot spots in the graphics, playing games, and engaging in other types of activities. Nonreading time ranged from 42.8% to 70.3% of the time, with an average across 12 programs of 64.6%. Given those results, it is not surprising that students as a group gained an average of only 2.4 words per program.

Instructional Interventions
In the second observational study, students interacted with the same types of talking storybook programs but increasing amounts of instructional support were imposed. The purpose of this support was to focus students' attention on the reading task in order to maximize opportunities for gaining reading skills.

One important aspect of the support conditions was to limit the amount of time students spent in nonreading activities. However, because these types of activities appear to be highly motivating to students, they were not totally eliminated. On each page of a storybook, students were allowed to access two hot spots in the graphics. Also, at the end of each instructional session, students were given 5 minutes of free time when they could reread the story, interact with hot spots, or play games or other activities.

Each of the six students in the study interacted with 4 talking storybook programs, the first two under moderate instructional support conditions and the second two under high instructional support conditions. In the moderate support condition, students were asked to read each page of the story aloud. In the high support condition, students were required to read each page aloud until they reached a criterion of 90% accuracy.

In both conditions, storybooks with two types of text interactivity were used. Stories with medium text interactivity allow students to hear the entire page read aloud. Those with high text interactivity allow student to hear not only the entire page but also individual words read aloud. These variations in text interactivity allowed students to experience four different levels of instructional support:
o Moderate support, whole page only: Students read each page aloud once.
o Moderate support, whole page and individual words: Students read each page aloud once, after reading 3 individual words on the page.
o High support, whole page only: Students read each page aloud to criterion.
o High support, whole page and individual words: Students read each page and 3 individual words per page to criterion.

Analyses of the data from this study are currently underway. However, preliminary results suggest that, as the level of support increases, students are more likely to show gains in reading skill. The table below presents the average number of words gained from pretest to posttest for the 6 students in this study. Clearly, there appears to be an increase in word recognition gains as the students progressed through the four levels of support. These results are even more interesting when it is recalled that students in unstructured interactions with the software gained an average of 2.4 words.

Condition; Type of Support; Page/Words ;Reading Gains
1 ; Moderate ; page ; 4.8 words
2 ; Moderate ; page , words ; 6.8 words
3 ; High ; page ; 8.8 words
4 ; High ; page, words ; 9.6 words

Ongoing Research
Two studies will be conducted during the 1998-99 school year. The first study will be a large-scale, year-long investigation of the effectiveness of representative types of hypermedia-based children's literature programs on the reading performance of students with learning disabilities. In order to examine the effectiveness of this body of software in realistic classroom situations, teachers will be trained to deliver the interventions to students with learning disabilities who they serve in their classrooms. At present, plans call for two types of software to be studied: talking storybook programs with high text interactivity and minimal extraneous games and activities, and talking storybook programs with medium text interactivity and embedded games and activities. A high support instructional protocol will be used, with some adaptations for classroom situations.

The second study to be conducted this year will be a smaller scale, observational investigation of the effectiveness of bilingual talking storybook programs for Spanish-speaking students with learning disabilities. Among the areas of interest in this research are teachers' and students' preferences for Spanish versus English versions of the programs, the ways in which students who are English language learners choose to interact with text supports in English and Spanish, and the magnitude of reading gains students make in reading both English and Spanish text.

At the conclusion of these studies, it should be possible to determine whether reading instruction with hypermedia-based children's literature is a promising approach for enhancing the reading performance of students with learning disabilities, both those who are fluent English speakers and those who are English language learners. In addition, it should be possible to offer teachers at least tentative recommendations to assist them in selecting the most effective types of software and the most useful instructional approaches for incorporating that software into classroom programs for students with learning disabilities.

Acknowledgment

The contents of this article were developed under a grant from the U. S. Department of Education. However, those contents do not necessarily represent the policy of the Department of Education, and you should not assume endorsement by the Federal Government.

References

• Armstrong, K., Brand, J., Glass, G., & Regan, L. (1995, October). Special software for special kids. Technology & Learning, 16, 56-61.
• Boone, R., Higgins, K., Falba, C., & Langley, W. (1993). Cooperative text: Reading and writing in a hypermedia environment. LD Forum, 18, 28-37.
• Bottge, B. A., & Hasselbring, T. S. (1993). A comparison of two approaches for teaching complex, authentic mathematics problems to adolescents in remedial math classes. Exceptional Children, 59, 556-566.
• Erickson, K., & Staples, A. (1995, October). Living Books and talking book strips: Interactive reading lessons for children with developmental disabilities. Paper presented at the Closing the Gap Microcomputer Technology in Special Education and Rehabilitation Conference, Minneapolis, MN.
• Lewis, R. B. (1998). Assistive technology and learning disabilities: Today's realities and tomorrow's promises. Journal of Learning Disabilities, 31, 16-25, 54.
• Lewis, R. B., & Doorlag, D. H. (1999). Teaching special students in general education classrooms (5th ed.). Upper Saddle River, NJ: Merrill.
• Lundberg, I. (1995). The computer as a tool of remediation in the education of students with learning disabilities--A theory-based approach. Learning Disability Quarterly, 18, 89-99.
• MacArthur, C. A., & Haynes, J. B. (1995). Student Assistant Learning from Text (SALT): A hypermedia reading aid. Journal of Learning Disabilities, 28, 150-159.
• Olofsson, Å. (1992). Can dyslexia be compensated for by computer-aided reading and synthetic speech? Fourth International ISAA Conference, Stockholm.
• Reid, R., Maag, J. W., & Vasa, S. F. (1993). Attention deficit hyperactivity disorder as a disability category: A critique. Exceptional Children, 60, 198-214.
• Ruiz, N. T. (1989). An optimal learning environment for Rosemary. Exceptional Children, 56, 130-144.
• Ruiz, N. T. (1995). The social construction of ability and disability: II. Optimal and at-risk lessons in a bilingual special education classroom. Journal of Learning Disabilities, 28, 491-502.
• Sindelar, P. T. (1987). Increasing reading fluency. Teaching Exceptional Children, 19(2), 59-60.
• Wise, B. W., & Olson, R. K. (1994). Computer speech and the remediation of reading and spelling problems. Journal of Special Education Technology, 12, 207-220.