Dr. Fletcher is Principal Investigator of the Texas Center for Learning Disabilities and is a former member of the NICHD National Advisory Council and the President’s Commission on Excellence in Special Education.
To learn more, visit Dr. Fletcher’s website at the University of Houston.
How many children struggle with dyslexia?
How many children struggle with dyslexia. You hear all kinds of figures from less than one to 20 to one in every five children. And the problem with estimates of prevalence is they depend on a threshold and the definition used for dyslexia. The most common figures people use are between three to seven percent of the population.
But that probably assumes relatively pure cases, no comorbidity, good instruction, things of that sort. There are many more children who aren’t called dyslexic who struggle to develop foundational reading skills.
Why is early intervention so important for kids with dyslexia?
Early intervention is very important because in many instances effective instruction can prevent dyslexia. It includes early screening because in order to identify children who are at risk we need to identify them early and introduce effective instruction at a time where we can really optimize a child’s access to print. It’s much better to intervene earlier than to wait until the child actually fails because remediation is demonstrably less effective.
And one recent study by Maureen Lovett, for example, she found that outcomes were almost twice as good if they were delivered in first and second grade and in third grade. We understand a lot about why that happens. In order to learn to read — because it’s an acquired skill — you have to give the brain an opportunity to learn about the relationship of what words look like and what words sound like.
Ultimately we want people to read at the whole word level. We want them to be lexical readers so that they recognize the whole word based on the statistical properties of the letter. To do that you have to access print early and you have to develop considerable exposure to print, and that allows the brain to program the systems that are needed for automatic reading.
If you don’t get early intervention and you wait to fail or if you get early intervention and you still struggle, those systems don’t develop well; and it’s very difficult to get the exposure needed to make those brain systems work.
What actually happens inside schools is that children are identified relatively late in their development. I think in special education, for example, the average age of identification is 10 years old. And many children are identified even later than that. It’s just the opposite of what we should be doing. We should be identifying children with reading and behavior problems as early as possible to prevent the cumulative problems that emerge, some of the ancillary difficulties such as anxiety.
One of the key attributes of children who don’t respond adequately to reading instruction is the development of anxiety.
What can we learn from screening kids early?
If I were the reading king I would have required mandatory, early screening for all children in kindergarten and grade one and grade two. We’ve had a law like that in Texas since 1997. And what we would do is develop, well, we have developed very short, for early screening we developed short instruments that take less than five minutes, they’re teacher administered. They vary depending on what the child needs to know in kindergarten, grade one, and grade two.
And they accurately identify kids who are not at risk. And that may seem odd to talk about identifying kids accurately who are not at risk, but a lot of people worry about over identification of kids at risk. It is far more serious an error to miss a kid who is at risk than to identify a kid not at risk, because you should follow up — an early screening instrument with a re-inventory or progress monitoring and that will tell you which child is going to be developing as they should and which child is still going to struggle.
So missing a child is a far more serious issue than falsely identifying a child. So our instrument is geared towards identifying children who are not at risk and very accurately the false positive rate goes down over time quite a bit, so that the instrument is quite accurate by the middle of first grade.
In Texas we developed the Texas Primary Reading Inventory for schools to use in Texas. It was free to schools in Texas and 1.95 percent of all the school districts in Texas were using it. We then expanded the list in Texas to include other instruments that showed reliability and validity. The idea behind the screen is that it has to reduce teacher burden. It has to be very short, it has to permit rapid triage of the entire classroom, and it does one simple thing in a very automatic way.
It either says that a child is at risk or not at risk. Once you identify children who are at risk, you spend more time doing an inventory or reading assessment or progress monitoring, something like that, because you need to put the child into a surveillance system where you know how all children are doing. And then you can begin to introduce interventions in the general education classroom and then intensify the interventions based on how the child responds to general education classroom instruction.
Why can’t special education do better helping kids with dyslexia?
Special education can’t fix all the literacy problems that children have because it’s a wait to fail system. It’s not oriented towards early intervention, and there are far too many children who need literacy intervention. The goal should be to reduce the number of kids who need special education services to a very small number so that you can give them very intensive interventions. The level of intensity that’s needed to remediate a reading problem is much more intense than people typically imagine.
How can we make early intervention a more common practice in schools?
To make early intervention a more common practice, we need to make the technology as easy as possible. You can support it with technology, with computers or what we used to call with iPads and things of that sort. It makes it easier for the teachers to administer it.
But the most important thing is to make it short and to reduce the burden on the teacher. When teachers see that it’s simple, that it’s easy to use, and they have the information about which children to worry about the most, it makes a difference.
What’s new in our understanding of what causes reading disability?
We have new understandings of what causes reading disability. And it’s clearly a combination of genetic factors that make the brain at risk and environmental factors that involve language in the home, exposure and things of that sort. Just imagine if you were a poor reader yourself, are you likely to read to your child when they’re really young? It’s hard to separate these different factors. And that’s why the concept of risk is so important in the disability world.
If we use the concept of risk and we think about prevention and we think about the fact that the neuro systems, in particular are malleable, that they change, that you can demonstrate that they change in development through intervention, we have a much broader understanding of reading problems and reading development in general that would really ameliorate these problems for many children.
What does effective intervention look like?
Effective instructional environments clearly can prevent learning disabilities, particularly when they’re done in kindergarten, grade one. You can even start in preschool. In preschool we should have oral language curricula include exercises that involved attention to parts of words, like rhyme, for example. If a child learns to listen to internal sounds that words would make — I’m not talking about teaching phonics to four year olds or anything of that sort — I think actually exposure to a broader oral language curriculum for all children, but particularly children who are at risk because they go to Head Start or Title One schools or things of that sort are particularly important.
In kindergarten the curriculum needs to include exposure to phonological awareness, which is an overarching understanding that words have internal structures that are related to sound. They need exercises and games that help them develop that skill. There’s good experimental evidence that shows that children who are exposed to phonological awareness teaching in kindergarten develop better reading outcomes in grades one and two.
By the time you get to grade one, you need good, strong core reading curriculum and teaching practices that allow teachers to differentiate instruction according to children’s individual strengths and weaknesses. So a good core curriculum includes explicit instruction that’s teacher led where the teacher models and demonstrates what the child needs to learn. It includes explicit instruction in phonics, which is the rules where letters and sounds are connected.
It doesn’t have to be drilling. So a it’s a mischaracterization to say that every child needs to be drilled in phonics, in isolation, because a good core curriculum includes a component to promote practice and opportunities to read, engagement. It should include comprehension exercises. So we talk about good strong curricula as including phonics, fluency, and comprehension that are integrated in a meaningful sort of way, not in a balanced way. That means phonics in isolation, comprehension instruction in isolation is much less effective than an integrated program.
And then for children who don’t respond adequately, particularly in first grade, a tutorial type of intervention. They need to spend 90 minutes a day in language arts instruction, and they need another 30 to 40 minutes in small group instruction. That is the most powerful way that I know to prevent a reading problem. In studies that we’ve done we’ve been able to reduce the number of children who are at risk from approximately 20 percent of first graders to two percent of first graders just through this tiered, layered intervention.
Strong core instruction for everyone, which absolutely reduces risk, and then you reduce risk even further by increasing time of task and increasing the amount of differentiation. And it incurs in the context of universal screening and progress monitoring. And then some children, and it should be a very small number of children if you have a strong core curriculum and effective tutorial interventions, they need some type of intensive intervention.
And that will often focus on a component skill that the child has not learned. It might focus, for example, on very intensive instruction in phonics. It could focus on very intensive instruction in fluency. It could be very intensive instruction in vocabulary if the child is able to decode but can’t comprehend. And those need to be probably an hour a day. Sometimes they need to extend for years. And if the child isn’t reading on grade level, that child has a persistent reading problem and that’s probably a disability at that point.
How does intervention impact the brain?
Well, instruction has very specific effects on the brain. We have brain systems that are evolutionarily designed for language and for visual attention processes, separate systems in the brain, and they operate in parallel. So when a child or a person for that matter, somebody who’s illiterate, is initially closed to print, and in particular to letter sounds in print, there are areas in the language areas of the brain that ordinarily process thematic code through the ears.
When we start using our eyes to process print, what the brain does is it engages these systems so that they become sensitive to constituent parts of words. We use a term like sub-lexical. Or you can just think about it as internal structures of words. Once that happens, and it has to happen, it’s the only way that you can ever learn to read.
You will never learn to read, for example, if I held flash cards and just flash them in your face, I have to tell you something about the internal structure of the word for your brain to engage, in order to engage this neural system that you need to support reading, because you’re not born with it.
Does multisensory instruction have particular value for children with dyslexia?
Multisensory instruction has a lot of different meanings, and it’s very difficult for people to specify what it means. In the older versions of reading interventions like Fernald, for example, and some of the early versions, what are now called Orton-Gillingham approaches to teaching, what it meant is teaching to all the senses, teaching to the auditory modal and to the visual modality and to the tactile modality.
To the degree that that type of instruction has been added onto what are called structured literary interventions, there’s not much experimental evidence that shows that that type of instruction is really important. In contemporary Orton Gillingham approaches, for example, they may have a multisensory component, but they also have a structured literary component that teaches about, teaches phonics basically.
And that component is very important. The multisensory component as traditionally defined does not appear to be effective, and it doesn’t seem to be important for children specifically for dealing with dyslexia. It’s probably better to think about multisensory as representing multiple modalities so the child sees a word, says a word, writes a word, uses multiple modalities, not sense modalities, but just different modalities to reinforce learning.
That’s a much better definition and it’s characteristic of many good instructional programs. If we restrict instructional programs for children with dyslexia to what are called Orton Gillingham or structured literacy or multisensory programs, many children are not going to respond just to those programs, and we need a broad array of appropriate reading programs for children to receive intervention.
How has your perspective on dyslexia changed over the years?
Early in my career I was very oriented towards children with disabilities and trying to give effective interventions for children who has severe problems who couldn’t read. My perspective has changed in two respects. First is the importance of early intervention, and the second is the importance of good general education instruction.
I’ve realized we cannot remediate our way into good literacy and that the only way to really deal with dyslexia is through a combination of general education and special education programs.
One reason is financial, because we need to provide more intensive intervention for those who struggle, and we don’t have the resources to do so if there are lots of children who need services. But the other reason is that it’s clearly more efficacious to provide instruction through the general education classroom than through supplemental kinds of interventions like special education or tutorial intervention and general education.
Outcomes are simply better because there’s more time on task and the child gets more of an opportunity to learn to read.
What is balanced literacy?
Well, “balanced literacy” is a term that we probably developed. Barbara Foorman and I actually used that as the title for a report that we did for the Houston Independent School District on effective reading instruction. And the idea was that phonics instruction could be balanced with literature-based instruction.
Fortunately what that turned out to be was a buzz word for literacy based instruction that’s instructivistic, holistic, isn’t very explicit. It’s an incidental phonics instruction. So now instead we talk about the importance of comprehensive instruction that integrates phonics fluency and comprehension.
The term “balanced literacy” has been hijacked; and if a parent hears that their school is using balanced literacy and she has concerns about a child’s reading ability, she should be, he or she should be very concerned.
How are background knowledge and vocabulary related? And how do they impact reading?
I think the biggest need is for explicit teaching of the alphabetic principle, explicit phonics. But the other thing, the other problem that many children with dyslexia have is more general difficulties with oral language. And that’s where teaching of background knowledge and vocabulary becomes especially important.
One of the really interesting things that we found in our research is that assessments of background knowledge and assessments of vocabulary are almost indistinguishable. They’re so highly correlated. People that have low background knowledge tend to have low vocabulary. People that have high vocabulary tend to have a lot of vocabulary background knowledge.
So teaching one or both or either seems to facilitate development of the other and certainly is linked to comprehension.
How does poverty contribute to reading difficulties?
Well, I think that poverty is a poor excuse for kids not learning to read. In our work we’ve focused on schools that have poverty characteristics. We’re not able to distinguish children who have problems with foundational reading skills and who are poor from children who are identified with dyslexia and are middle class, except possibly in their oral language characteristics.
The types of teaching methods that work with economically disadvantaged kids also work with kids that have dyslexia and vice versa. That’s a real misconception that different kinds of teaching methods are needed. The other misconception is that the predominant problems in children who are economically disadvantaged are comprehension and language development more broadly.
In one study that we did in middle schoolers who didn’t pass the state reading comprehension test, we found that 81 percent of the children who did not pass the comprehension test had problems with foundational reading skills. In fact, they were at such a low level that we had to modify the lesson plans in order to provide more opportunities for word work.
Sometimes the problem — even though people talk about comprehension problems in one public report representing 70 percent of children who struggle at the adolescent or secondary school level — that’s not been our experience. Our experience is that it’s more like one in five, where the other 80 percent are still struggling with foundational reading skills, either accuracy of decoding or fluency of reading text.
What are some intervention fads that don’t help children with dyslexia?
I think there’s a tendency to look for shortcuts in trying to help children with dyslexia. And the types of shortcuts that people look for are often faddish interventions that don’t work. So one of the things that I tell principals, teachers, and parents is that there is no such thing as an effective intervention for dyslexia that doesn’t involve access to text. It has to be in the context of reading. So I discourage visits to optometrists or eye exercises. I don’t encourage brain training programs.
I don’t encourage exercise. None of those show any beneficial effect on reading problems. Even something as rudimentary as phonological awareness, and we all know that phonological awareness is very important to learn to read, but if you don’t have a letter component where the child learns to transfer what they’ve learned about phonological awareness in oral language to print it doesn’t generalize to better reading.
So any effective intervention has to have a reading and writing component. And if it doesn’t, it isn’t worth doing and it isn’t related to dyslexia.
How can technology interventions help children with dyslexia?
If the technology has a reading and writing component. it has a potential to be effective. If it’s training, for example, auditory processing skills, or brain games, like Lumosity, or Fast Forword, you know, slows down speech, there is no effect on reading. It has to have a text component. And I think it can be very useful, particularly for practice and things of that sort.
The extent to which we can teach people about word sounds and internal structures of words with computers is questionable to me. I still wonder if we don’t need a human to provide feedback. I certainly think we can use computers, hypertext to encourage kids to read more and to spend more time using text. I think you can do comprehension lessons on computers. It’s the word-level work that concerns me the most.
Why hasn’t what researchers have learned about teaching reading changed what teachers do in schools?
The biggest problem that we have is an absolute failure to translate what we know from science into schools. There’s a big epistemological gulf between what we know about reading research and how that gets implemented in schools. In many areas such as teacher preparation, there’s an outright rejection of science as a way of knowing about what’s effective and what’s likely to work. Until we bridge that epistemological gap, we’re going to continue to have translational problems.
I don’t understand why people aren’t more accepting of science, particularly at the level of curriculum instruction. It would seem an obvious thing to me, but for many years, anecdotes, personal experiences, odd ideas about what’s good for children have prevailed. And it’s been very difficult to penetrate that.
I also think there are economic reasons behind it, because I think that there are certain approaches of teaching to read that are very lucrative and fairly easy for the progenitors to produce and sell, even in a lack of efficacy or alignment, even content alignment with what we know from research.
Well, guided reading and Reading Recovery stand out to me. I mean they certainly help some children, but they don’t help children with dyslexia.
What does it take for a school to embrace the science of teaching reading?
I think that when schools have the opportunity to see reading practices that work and when you make it possible for them to see actual changes and gains in the student population and show them that you’re reducing referrals, for example, to special education, I think you see a lot of the teachers rallying around that, particularly if the leadership is behind it.