Angela Swift had been a special education teacher for years when her kindergarten-age son, Toby, was diagnosed with dyslexia. The La Veme, California, public school the boy attended gave him extra help to compensate for the problem, but nothing seemed to work. By 3rd grade, Toby was bright enough to score A’s on oral exams, but he still could not read. Swift began to search for other ways to help her son. She soon discovered a new body of research on reading disabilities, much of it pointing to strategies that just might help Toby. “I was amazed,” she recalls. “How could I be around special education for 15 years and not know about this? How could my school district not know?”
Most of this new research has been supported by the National Institutes of Child Health and Human Development, an arm of the National Institutes of Health in Bethesda, Maryland. It has helped dispel some common myths about dyslexia. It also has pinpointed a possible biological origin for reading disorders and identified ways educators can spot and treat such problems before young children fail in school. But the work has been controversial, in part because it taps into the national debate between whole language proponents and phonics advocates over the best way to teach reading. As a result, schools and teacher training programs have been slow to embrace the findings.
Traditionally, children have been labeled dyslexic if they are otherwise intelligent and suffer no other physiological problem-a hearing impairment, for example-that might cause reading difficulties. But many researchers now believe this definition is too narrow. They maintain it leaves out large numbers of children who cannot read because they are from disadvantaged homes and haven’t been exposed to books, read to, or taught the alphabet. Dyslexia, one such researcher says, “is just a fancy name for not being able to read”.
Because of the disagreement over definitions, estimates on the number of dyslexia children vary wildly. Some researchers put it as low as 5 percent. Dr. Sally Shaywitz, a researcher at Yale University’s school of medicine, has tracked 451 schoolchildren in New Haven, Connecticut, since 1983 and estimates that some 20 percent are reading disabled. G. Reid Lyon, who heads up the NIH’s division on child development and behavior, says the slice could be much larger-as high as 40 percent-if you include children who can’t read because they were not exposed to the printed word or were not taught well.
Contrary to popular belief, people who are dyslexic do not necessarily see or write letters backward or in reverse order. They just can’t make much sense of the written language. As Swift’s young son Toby puts it words “just look like a lot of letters.”
In fact, the latest research suggests that reading problems have little to do with the print a child sees on a page. Rather they may be a manifestation of a much broader problem with language comprehension itself. Someone who has trouble reading, for example, might also have problems with word recall and idiomatic expressions.
Virtually all studies of dyslexia agree on one point: The problem is rooted in a person’s inability to distinguish phonemes, the smallest sounds that make up speech. The word “bat,” for example, is made up of three phonemes, or sounds. The English language contains 44 such sounds in all.
Many people with dyslexia can’t detect the tiny sounds in “bat,” Lyon says. “The minute I say that word, the ‘a’ and the ‘t’ fold up into the ‘b’ sound and become one pulse of sound.” In normally developing readers, the brain registers those constituent sounds. With this skill in place, beginning readers can then begin learning phonics, the process of linking letters to sounds.
Those who cannot distinguish the small sounds will have little success decoding words phonetically. The sounding-out process is so laborious that they have no energy left to understand what they have just read. Therefore, children with reading disorders may need to be trained to distinguish the phonemes.
Frank Vellutino, a professor of psychology at State University of New York at Albany, suggests a simple diagnostic tool. “Give children a set of nonsense words,” he says. “The child who can’t pronounce them is the child with basic reading disabilities.” In his studies, Vellutino has been able to identify potential poor readers as early as kindergarten by asking them to perform small tasks, such as pronouncing nonsense words, identifying letters of the alphabet, and distinguishing sounds that make up words.
The notion that reading disabilities and language problems might have deeper biological causes has been bolstered by NIH research on human brains. In one such study, Dr. Albert Galaburda of Beth Israel Deaconess Medical Center in Boston examined autopsied brains of poor readers and found them to be fundamentally different from those of normal readers. Some neurons or brain cells form out of place in the areas of the brain that control language-an abnormality that occurs before a child is born. The defect, he says, may affect the way nerve cells connect with one another later in life, as well.
The research that most directly bears on the futures of poor readers like Toby Swift was conducted in the classroom. Vellutino and colleagues at SUNY-Albany worked with 76 middle-class children with reading problems who otherwise showed average or above-average intelligence. Beginning in 1st grade, the students were given 30 minutes a day of one-on-one tutoring. During this time, they read new stories and familiar ones, studied letter-sound relationships and useful spelling patterns, and learned to recognize a number of words by sight.
Vellutino’s aim was not to pinpoint what specific strategies worked best but to find out whether the tutoring sessions could be used as a means of diagnosing readers most in need of help. He discovered, however, that the diagnostic sessions improved the reading of almost all the students.
At the end of one semester, 70 percent of the children were reading in the average range. At the end of two semesters, only 15 percent of the students were still struggling.
Joseph Torgeson, a Florida State University psychology professor, undertook a similar experiment with 60 4th graders ranked among the bottom 2 percent of readers in their grade level. The students were divided into two groups. One group was taught “embedded phonics” and practiced letter-sound relationships and did extensive writing. The other was taught using a method known as “auditory discrimination in depth.” This approach essentially teaches children to feel the sounds in words; it emphasizes the shape of the mouth and position of the tongue when pronouncing a letter sound.
Both groups of children received two hours of individual lessons a day. After eight weeks, each group as a whole had moved up into the average reading range-an improvement of one and a half
to two grade levels. But the auditory-discrimination technique proved more successful than the other. Only 9 percent of the children taught that way continued to have serious problems, compared with 26 percent of the children taught embedded phonics.
“These gains are strikingly different than what is typically obtained in the regular resource room,” Torgeson says. “People are afraid of one-on-one instruction because it’s very expensive,” he adds. “But our rationale is, well, you need to know what you can accomplish. Why take two years when you can do it in eight weeks?”‘
None of the strategies in the new studies worked for all students. Dyslexia is a lifelong affliction, and some researchers estimate that roughly 5 percent of children may never learn ways to compensate for it. But the gains the studies document represent, for the most part, big improvements over typical practice. What’s more, most of the research points to a clear strategy for teaching reading disabled children. It suggests that at a minimum poor readers need explicit instruction in phonemic awareness, phonics, and the structure of language. And they need enough practice to enable them to use those skills automatically.
The problem, the researchers say, is that over the past decade schools have moved in the opposite direction-away from phonics-based approaches and toward whole language methods of teaching reading. Proponents of the latter maintain that children learn to read naturally and call for immersing them in good literature rather than putting them through repetitious phonics drills. Although whole language educators often include phonics in their lessons, they do so in the context of stories children read.
What we need, Vellutino says, is “a diverse set of strategies and a balanced approach to reading instruction.” In arguing for balance, Vellutino isn’t just talking about children with reading disabilities. Along with Lyon of NIH and other researchers, he believes that all young children would benefit from an approach that includes phonemic awareness, phonics, engaging literature, and lots of writing. They suggest that learning to sound out words may be the gateway children must cross if they are to read at higher levels.
As they’ve lobbied for this approach around the country, however, such ideas haven’t always been well received, particularly within the education research community. Richard Allington, an education professor and the chairman of the reading department at SUNY-Albany, says the researchers may be overreaching by broadly applying to all classrooms their work with poor readers. Many of the gains documented by Vellutino and other NIH-supported researchers, he argues, have come on sound-awareness tasks or exercises that ask students to read nonsense words. “They don’t have reliable results on real-word reading, fluency, and comprehension,” he says. “The last time I looked, no one had targeted children with disabilities’ primary need as being able to pronounce nonwords.”
Meanwhile, additional NIH-supported research on reading disabilities is on the way. Studies begun this fall are using magnetic resonance imaging to look into the brains of children who are poor readers – both before and after they have undergone intensive tutoring. Previous studies using the technique, which is harmless to the kids, have already suggested that the metabolic activity in the brain of a dyslexic child trying to read is different from that of other children. Now, researchers are looking to see whether successful reading interventions can change those brain-activity patterns.
After learning about the NIH-supported research, Angela Swift began wrangling with the Bonita Unified School District, where her son goes to school. She wanted the district to introduce some of the new reading strategies into Toby’s education. It was an uphill struggle.
That wouldn’t surprise Louisa Moats, a clinical assistant professor of pediatrics at the Houston Health Science Center at the University of Texas. As part of her research, Moats reviewed four of the most popular textbooks used in reading-education courses for aspiring teachers. “None of the texts contained accurate information about phonology and its role in reading development,” she told a congressional committee this year. “And none of them explained with depth or clarity why many children have trouble learning to read or what to do about it.”
Among the hundreds of teachers Moats has informally surveyed, fewer than 10 percent could name major contributors to the field of reading research. “It was astonishing to me.” she says, “the ignorance of the field about this.” Frustrated by their son’s lack of progress, Swift and her husband decided last year to hire a private tutor for Toby, who has been formally identified as gifted. The tutor used the same auditory-discrimination technique that proved so successful in Torgeson’s study. It helped, Swift says.
Finally, after months of pressure from the Swifts, the district agreed to train several teachers, including Toby’s, in the technique. It also formed a committee to explore other methods that might help struggling readers and to create a reading curriculum for special education students.
Originally in Teacher Magazine, January, 1998