CAST: Universal Design for Learning

At CAST (Center for Applied Special Technology), a four-year-old girl sits at a computer with her mother, exploring Just Grandma and Me, an electronic storybook. When she clicks on a word, the computer says it aloud. The child repeats the word. Sometimes she says the word along with the computer, not after. After clicking through the text one word at a time, she says the whole sentence aloud, turns to her mother, and exclaims, "I can read!"

At the Harvard Literacy Lab, students with reading and writing difficulties willingly compose autobiographies, stories, and poems for personal home pages posted on the Lab's Web site. When e-mail responses to their writing arrive from other parts of the country and from as far away as Japan and Australia, they proudly mark the location of the sender on a map of the world.

These two examples illustrate how computers are changing the way children are learning to read and write. Some educators find these anecdotes exciting. They show the potential of new technology to revitalize reading instruction and to make reading more relevant to the lives of children growing up in the electronic age. Other educators find these anecdotes troubling. The first anecdote sounds more like "word-calling" than reading, and the second seems almost "anti-literacy" — one more example of the erosion of traditional literacy in our culture.

For the majority of reading teachers, we suspect, these anecdotes raise important questions about priorities and resources. Given the limited time available in the classroom for teaching reading, how valuable are computer-based activities compared with other learning activities? Given limited computer resources, are these the most effective ways to use technology?

We believe computers can play an important role in literacy development, but considerable care must be taken to identify what that role is. Not all literacy software nor all strategies for using the computer to teach reading are valuable enough to consume limited time and technology resources.

Although we cannot resolve all of these concerns, we can begin to address them by examining computer technology in the context of new research on reading and learning. Bringing these two kinds of understanding together — considering first how reading and learning actually happen and then what computers can do — provides a solid foundation for determining how technology can effectively contribute to the processes of learning to read and write. Although we use specific literacy applications as examples, we have chosen not to provide a guide to good and bad software currently on the market. The roster of software changes so rapidly that such a guide would immediately be outdated. Instead we provide guidelines for evaluating technology products and teaching strategies — guidelines that should remain current despite technological change.

Posting this book on-line in a digital format increases accessibility for readers with a variety of individual needs and purposes. For example, anyone can search the text, customize its appearance, or have it read aloud. Images can be turned on or off, accessed through image descriptions, or customized in appearance. Availability on the Web also provides direct access to Web sites mentioned in the book. Additionally, images on-line can be presented in color whereas the book's printed images are grayscale only.

How Do We Learn to Read?

Introduction

Recent studies of how the brain functions show that learning to read is a complex, individual process that requires multiple skills. Popular literature often misinterprets reading as primarily a matter of word recognition, to be learned in the first years of school. The reality is different. Chall (1996) identifies six stages in the process of learning to read; many people never reach the highest stage, and those who do typically take 20 years to get there. Teachers know that reading instruction extends well beyond elementary school and includes the development of deep content knowledge, critical evaluation of information and ideas, and building lifelong interest in learning (Stahl, Hynd, Glynn, & Carr, 1996).

Some advocates of a "balanced reading program" suggest that phonics instruction and whole language approaches should be combined in varying proportions to reach the largest number of students (Adams, 1990; Clay, 1991). We agree that a balanced program is desirable, but focusing on a dichotomy between "whole language" and "phonics" is not likely to achieve it. In fact, we believe that any dichotomy obscures what teaching reading, with or without technology, really requires. To get beyond single-minded or even double-minded solutions, we need a framework that supports a truly balanced approach to reading — one that recognizes the full breadth of the challenge and the whole range of individual differences in the students we are trying to teach.

Recent neurological research sheds light on the process of learning to read and can help shape the framework. Scientists now use PET scans and other advanced imaging techniques to study the living, working brain. Their efforts are giving us fascinating glimpses into how the brain works during reading, and provide insight into how a balanced approach to the process of learning to read should be designed.

Reading and the Brain

PET scans generate images of the brain that distinguish highly active regions from those that are less active. The more active a region is, the more glucose it metabolizes — creating a "hot spot" of energy consumption. The greater the activity, the more intense the hot spot, and the more brightly colored its appearance on the PET scan.

Findings on brain activity during reading have surprised more neuroscientists than reading specialists. PET scans show that there is no "reading center" in the brain, no one place where reading occurs. Several different parts of the brain are involved in reading, each making its own contribution to success. Knowing which parts of the brain "read" together and what their individual functions are tells us what skills and processes reading instruction needs to address. As we will make clear, it also suggests how technology can help.

The PET scans in Figure 1-1 show brain function associated with hearing and seeing words. Hearing words creates hot spots mainly in the temporal lobe of the cortex. Seeing words mainly involves areas in the occipital lobes. The dramatic differences in brain function during these two activities begin to suggest how complex learning to read is. The scans also imply something else. The hot spots form not just a pattern, but a network, because the active parts of the brain work together to perform what may superficially seem a simple task.

d

d

FIGURE 1-1. Adaptation reprinted with permission from Nature (see Petersen, Fox, Posner, Mintun, & Raichle, 1988). © 1988 by MacMillan Magazines Limited.

What parts of the brain burn glucose most intensely when someone reads? The answer depends on how skilled the reader is, the difficulty of the text, the goal of the reading activity (for instance, understanding main points versus searching for details), the level of motivation, whether the reader has a specific reading disability, and other factors. These differences are critical to effective teaching and to our discussion of the role of technology in the process. They show us why different learning tasks and individual learners require different approaches and materials. Equally important is the fact that reading is a whole-brain activity. Neuroimaging research has shown that many parts of the brain comprise the brain's reading network: each major area plays a different role, bringing a different "skill" to the cooperative process.

Recent neurological findings (Cytowic, 1996) confirm earlier descriptions (Luria, 1973) of three spatially and functionally distinguishable but interconnected systems in the learning brain. Broadly speaking, one system recognizes patterns, one generates patterns, and the third determines priorities. All three systems are crucial to reading and to learning to read. A brief description of how they work will provide the scientific grounding we need to understand how technology can help teachers teach reading and students learn to read.

1. Recognition Systems

Most of the posterior (back) half of the brain's cortex is devoted to recognizing patterns. Bruner's distinction between "knowing how" and "knowing that" (1976) helps clarify the difference between recognizing patterns (knowing that) and generating the patterns that direct action (knowing how). The recognition systems make it possible to know that a particular stimulus is a cup, your grandmother's voice, the smell of coffee—or the letter A, the word cat, a paragraph, or a chapter heading.

Damage to the posterior cortex can affect the brain's capacity to know what things are. Depending on the degree and kind of damage, an individual may lose the ability to recognize objects by their color or shape, by the way they move, or by the way they sound. Differences in the networks allocated to recognition in undamaged brains account for individual variations in the ability to recognize pitch, timing, location, color, orientation, or shape.

Many aspects of reading depend on pattern recognition: knowing that the letters c-a-t make a pattern that stands for the word cat; recognizing the silent e pattern; identifying a poem as a sonnet or a particular arrangement of words as William Faulkner's style.

2. Strategic Systems

The networks critical for knowing how to do things — taking steps, saying a word, shooting a foul shot, reading a book, driving a car, planning a picnic, writing a narrative — are located in the anterior (front) part of the brain. All of these activities are patterns of actions; all of our skills, strategies, and plans are essentially highly patterned activity. The front part of the brain generates patterns that underlie our actions, including the action known as planning.

Damage to some parts of the frontal lobes means paralysis; damage to other parts makes it difficult or impossible to coordinate movements into higher-level skills. Damage to the most anterior networks affects the ability to plan and organize long-term goal-directed action; they make individuals seem disorganized, impulsive, uncoordinated. Again, individual differences in undamaged brains lead to functional differences such as varying levels of fine motor skills and other kinds of physical and mental coordination, or different capacities for planning and organization.

Reading depends on pattern generation, including patterns of movement. Holding a book and turning its pages is a complex combination of skills that young children perfect through practice. The eye movements across the page that allow a reader to identify letters and words and turn them into meaning are acts of physical coordination carried out by the frontal lobes (Pollatsek & Rayner, 1990). The recognition systems cannot get the information they need unless the strategic systems know how to search the text skillfully.

The networks in the brain that build strategies are also essential to text comprehension. Understanding is more than reception and perception. Good readers approach text strategically. They monitor their own performance by making and testing predictions; they search in the important places of the text, scanning for salient information; they identify the organization of a story and draw inferences about meaning; they look at the parts of words to distinguish them lexically; they reread puzzling sentences (Anderson, Hiebert, Scott, & Wilkinson, 1985; Richek, List, & Lerner, 1989). All of these acts rely on the strategic, pattern-generating systems in the brain.

3. Affective Systems

The networks critical for feeling — the fear of heights, the desire for sugar, the sadness of loss, the excitement of hope — lie at the core of the brain, functionally connected to the limbic system. These networks neither recognize nor generate patterns — they determine whether the patterns we perceive matter to us, and help us decide which actions and strategies to pursue. The recognition systems can identify a certain combination of shapes, colors, and smells as a hamburger. Our desire and decision to eat the hamburger comes from our affective systems, which motivate the reaching, grasping, biting, and chewing actions that our strategic systems manage.

If we lacked the ability to establish priorities, select what we value or want, focus attention, or choose actions, our complex nervous systems would drown in a sea of patterns. Recent neurological work shows that human intelligence also depends on the capacity to determine which patterns count (Cytowic, 1996; Damasio, 1994; Goleman, 1995; LeDoux, 1996).

Without affective systems, readers could not direct and sustain attention to text in the face of millions of competing stimuli. The affective systems make it possible to look at letters on a page instead of birds outside the window or to finish an assigned chapter before eating. They enable readers to focus on words carrying information they are specifically seeking or locate the section of text that relates to their hypotheses. Individuals with impaired affective systems are likely to be poor readers for many reasons. They have difficulty focusing appropriately and establishing priorities for accomplishing reading tasks; they don't vary their rate or style of reading to match differences in content or purpose; they can't concentrate; they lack motivation to read or engage in reading-related tasks.

Individual Differences

The neurological research highlighted here confirms two truths important to the teaching of reading and to the role computers can play in the process. One, demonstrated by the participation of recognition, strategic, and affective systems in the process, is that reading is a complex activity, involving the coordination of many skills. The other is that no two brains work in exactly the same way. While everyone's brain functions take place in roughly the same areas and work together in roughly the same way, PET scans show that each individual has his or her own activity "signature." Each of us has a different functional allocation of cortex. Some people have larger regions devoted to recognizing patterns, generating strategies, or focusing on particular priorities and these differences seem to be reflected in different configurations of ability.

This finding probably surprises more neuroscientists than reading teachers, who have learned from experience that students have different learning styles, relative strengths and weaknesses, and varying kinds of intelligence (Carbo, Dunn, & Dunn, 1986; Gardner, 1983). The fact that individual students develop and coordinate the many skills involved in reading proficiently at different rates and in different ways helps explain why teaching reading is so demanding, so much an art as well as a science. The challenge is increased by limited time, large classes, practical difficulties in providing individually appropriate reading materials to individual learners, as well as finding approaches that stimulate and support a broad range of learners.

What Are Computers Really Good For?

It takes time to figure out how to use a new technology — to discover the valuable new uses implicit in the technology itself. At first, people tend to use new devices as if they were just different versions of something older and more familiar. When Marconi invented the radio — which he called the "wireless telegraph" — he thought it would be used like the telegraph to communicate important messages between two points. Only later did it become clear that the wirelessness of the new technology was its most important feature, giving it the previously unimagined power to broadcast messages (as well as music, comedy, and drama) from a single source to many listeners.

Likewise, we are still in the process of discovering what roles computers will play in our lives. When they were huge, expensive number crunchers, some experts predicted that five computers would take care of the world's need for specialized military and scientific calculation. By the mid-1980s, powerful small computers sat on millions of desks in businesses and schools, performing an array of tasks including accounting, word processing, publishing, graphic design, and learning simulations. At the end of the 1990s, multimedia capabilities have vastly increased computers' ability to entertain, inform, and educate. The World Wide Web has also turned them into extraordinary communication tools.

Taking into account our incomplete understanding of computers and ignoring advertising claims for their glitzy features, can we identify some basic quality that differentiates computers from other technologies and gives them the power to contribute to the teaching of reading? In our view, that one quality is flexibility. More than any other invention, computers are multipurpose devices that perform different functions, essentially becoming different machines through the use of varied software. This unique flexibility is the source of two related strengths.

First, computers are versatile. They can emulate a book, an audio CD player, a video game, a telephone, a VCR, a spreadsheet, a drafting table, a musical instrument, an editing studio, or even a battlefield. Using a computer, students can write, draw, compose music, animate pictures, ask a word or letter to say its own name, have a story read, see the lyrics to a song while listening to it, redraw pictures to suit their fancy, or record their own voices to accompany their own text or pictures. No other technology approaches this kind of versatility.

Second, computers can be customized. Adaptable to many tasks, they can also be adapted to many users. Printed classroom materials come in one format; one size fits (and often fails to fit) all students. Computer-based materials on the other hand can be adjusted to meet the particular needs of students who vary in the strengths and limitations of their sensory, motor, cognitive, motivational, and emotional makeup, their exposure to literacy, their language and cultural backgrounds, and their stylistic preferences. Our brief discussion of neurological research shows how profound some of these individual differences can be, and how important addressing them is to successful reading instruction.

The process of using the flexibility of computers to build customizing options for individual students is one aspect of universal design. For ten years, CAST has conducted research on universal design. Our initial goal was to use computers to make curricula more accessible and useful to students with disabilities, but over time we have learned that this same flexibility can benefit all students. Learning is a complex and idiosyncratic process. Because computers are flexible enough to do many different things in many different ways, they can help us create appropriate learning environments for every student.

Universal design offers varied representations of information to adjust to the recognition needs of all students, including children with learning disabilities, visual or auditory impairments, physical disabilities, and diverse learning preferences. For example, reading material can be presented in the following ways:

• with variable text size
• with variable text, background, and highlight color
• read aloud in synthetic or digital speech, with or without music
• with images described in words rather than presented visually
• through animations that match text chunks, reflecting text meaning directly
• with phonemes, words, phrases, or sentences read aloud on request
• through a video of someone using sign language to translate the text or audio track
• through combinations of the above options

Universal design also offers varied options for expression, supporting students' individually different motor and strategic systems. For example, students can be offered:

• options to respond through text, recorded speech, images, or video
• supports for spelling and typing such as word prediction software and voice recognition systems
• opportunities to explore text and images by manipulating them
• optional supports such as think-alouds, leading questions, suggested strategies, and templates.

Universal design also supports different affective systems, adjusting to students' different interests and needs for challenge, support, and novelty. For example, computers can offer students the following:

• options to choose their own material and tailor activities to their needs, preferences, and skill levels
• opportunities to read and write in real-life contexts
• adjustable challenges and supports
• tools and resources to create original work in many modes and styles
• timely and appropriate feedback

A universally designed computer reading program lets children with differing visual abilities and preferences choose text sizes, colors, and fonts that best meet their individual needs. They can turn "pages" and move through activities in ways that suit their various motor abilities: pointing to the screen, using the keyboard or mouse, even pressing a single switch to select among options that highlight sequentially on the screen. Some programs let users write and draw in similarly diverse modes. In addition to giving students ways to adjust the learning environment, these programs offer a range of reading supports. Students can get as much or as little help as they need, clicking on words to hear them pronounced or defined, seeing words sequentially highlighted to support left-to-right tracking, selecting annotations or animations that give clues to meaning, among other options. Printed text offers none of this flexibility. In fact, print forces students to try to fit themselves to a single, rigid "standard," impeding learning for many. Universally designed programs offer many sizes. Their universality comes from flexibility, not an assumption of sameness.

Computers help people learn difficult tasks by simulating the tasks in safe, supported, guided ways. Flight simulators use interactive multimedia to create the look and feel of flying an airplane by changing the world displayed on its window-monitors in response to pilot actions. Novice pilots repeatedly simulate takeoffs and landings, mastering the basics through extensive, safe practice. In the reading classroom, computers can similarly function as "reading simulators," providing appropriate learning challenges along with guidance, support, and an opportunity to practice. A computer environment that offers guidance and feedback without impatience is an important contribution to the learning process.

Predictions about new technology in the reading classroom are much like those about other innovations. Some consider computers a passing fad, unrelated to the primary task of teaching reading. Others alarmingly predict that technology will dominate teaching, ultimately replacing teachers entirely. Still others expect that the new media will supersede printed books as the primary means of encoding information in our culture (Gilster, 1997; Nunberg, 1996) and that a different kind of literacy — electronic literacy — will emerge.

We believe that computers should and will play a major role in the reading classroom but will almost certainly not replace books or teachers. They will influence and perhaps even redefine traditional books, literacy, and the role of teachers, but all three will survive and thrive. We believe too that developing a clear-sighted, open-minded understanding of both old and new technologies will help develop a complementary relationship between them.

Laying the Foundation

We have described characteristics of reading and learning processes and the characteristics of computer technology that we believe should inform the evaluation of reading technologies and strategies. These are the essential points:

• Reading and learning to read are complex, whole-brain processes requiring the use and coordination of a large set of skills.
• Individuals learn to read in different ways and at different rates, and effective instruction must attempt to respond to these differences.
• Practice, motivation, and support are essential to reading success for all students — and most critical to those with reading difficulties.
• Flexibility is the most important attribute of computers in the classroom, suiting them to the complexity and individuality of the learning process.
• Computers can be adapted to present material in many ways and customized to individual learning styles and needs.
• The interactive quality of computers allows them to engage, motivate, guide, and support students.
• Computer technology is likely to complement rather than compete with books and teachers.

In the chapters that follow, we discuss the ways electronic technology can support recognition, strategy-building, and motivation — the three critical components of a new view of balanced reading instruction. We will describe examples of currently available applications and consider some future improvements that can make computer software an even more powerful instructional tool.