The Gaab Lab explores questions related to learning disabilities and especially typical and atypical language and reading development. Current research is looking at neural pre-markers of dyslexia in infants and preschoolers, brain correlates of reading fluency, and potential connections between musical training and language and reading development.
What is the dyslexia paradox?
The dyslexia paradox is the discrepancy between when we currently diagnose dyslexia and when research has shown the most optimal window for early reading intervention is. So currently we are diagnosing kids after repeated failure — so we also call it the “wait to fail paradox” or “wait to fail approach” — which is usually at the end of second grade at the earliest, maybe beginning of third grade.
But then many children are only diagnosed way later than this. However, research has shown from several wonderful research labs that the most optimal window for early intervention is kindergarten and first grade — and most likely before that. But to my knowledge, there has not been any research yet. So that means we can’t get a diagnosis or identification often in the school district until late elementary school or middle elementary school, but the window is much earlier. So that’s what we term the dyslexia paradox.
What are the effects of delaying identification of and intervention for dyslexia?
So we know that dyslexia goes along with a lot of social and psychological and clinical implications such as reduced self-esteem, depression, higher rates of anxiety, reading motivation goes down, et cetera. So the longer we let the kids fail and not give them the resources they need, the more likely they are to have reduced reading motivation.
We know that they often show low self-esteem really, really early, and we also know that the gap between the typical developing children and the children who struggle with reading widens over time — what is often called the Matthew Effect of Reading.
Usually children start kindergarten with a vocabulary maybe around 10,000 words. At fifth grade these kids really increase their vocabulary to maybe around 60, 70,000 words. Before kindergarten these words are primarily learned through the social interactions between the parents, the caregivers, and the child.
We know that the increase in vocabulary between kindergarten and fifth grade is more — comes really from more reading, like reading — and you get the vocabulary from the text you read. So if you are a struggling reader, you read a lot less, which then leads to reduced vocabulary, which then just adds up to all the problems and all the symptoms the children show, which then can lead to problems with syntax comprehension, text passage comprehension, et cetera.
What does the research say about screening young children for dyslexia?
I think it has to be early. A lot of people think that early identification means that we move instead from third grade to second grade, which is not early in our opinion. So research has shown that interventions are most effective in kindergarten and first grade. So we want to know who needs these interventions, and so we need to identify children at risk in preschool or very early in kindergarten.
Also, it’s important to say that we’re not diagnosing or identifying children with dyslexia at that age. We are identifying children who are at a heightened risk to develop dyslexia, okay? So this is a risk evaluation, almost similar to evaluating you for being at risk for heart disease. We are not giving you a diagnosis of heart disease. We are just saying, “Oh, you have higher cholesterol. You are at heightened risk to develop heart disease.”
How can early detection of dyslexia be improved?
We also think that the screening methods really need to be improved. So currently there is not enough screeners out there. A lot of the screeners out there are not screening all of the components important for learning to read — meaning things like phonological awareness, which is the ability to manipulate the sounds of language, letter sound, letter name knowledge, et cetera.
There’s also the piece of oral language, which is really important for reading fluency and reading comprehension. So you want to also screen for things like vocabulary as well as oral listening comprehension, which is just listening to sentences and then deciding which picture goes with the sentence, which really illustrates whether the child has the capacity to understand complex syntax.
How do we select an effective screening tool?
So there is important factors to consider when selecting the right screener to identify children in your school classroom or district — and I’m going to say a list here. So first of all you want to make sure that it assesses all important skills that we know are important for learning to read.
Secondly, you want to make sure that the screener is evidence-based — that there’s some research behind it, that it’s not just someone who put some words together and said, “This is a great screener because it has worked last year in my classroom.” And what are the characteristics of the norming group? So I give you an example.
So if you are in an urban school, with many children from maybe low socioeconomic status families, you want to make sure that this was represented when the screener was normed, meaning: were kids included from lower SES families when they were norming the screener?
You also want to make sure that the overall time that it takes for you to screen the children is — works in your classroom, right? There is very short screeners. There are very long screeners. So you want to make sure that this works with your school and it works with your — what we call a screening protocol. So you could screen the summer before kindergarten — which may give you a little bit more time because you get the kids in individually.
Or do you need to screen within the first week of kindergarten and you have very limited time? You want to look at sensitivity and specificity of a screener. So sensitivity is correctly identifying those who will develop a disability, and specificity is correctly identifying those who will not develop a reading disability. So you can also call it the false positives and false negatives.
You have to make a decision how to deal with what we call false negatives and false positives. So false positive is, a child is identified as being at risk but is not actually being at risk, which can be due to problems with the instrument, the day they were tested they were tired, didn’t have breakfast, et cetera.
Then you have false negatives, which is — a kid is actually at risk, but you are not detecting it with your instrument. So as a district, we recommend that you catch all the kids who are at risk so you would have no false negatives, but that could be – that could lead to more false positives.
And so you can move these two bars and decide based on the resources, the number of testers you have, et cetera, how you want to move the discrepancy between the false positive and the false negatives.
Can brain imaging help with early identification of dyslexia?
So we do a lot of brain imaging in my laboratory, and at this present time, we do not think that brain imaging can be helpful for the early detection of children at risk. We also don’t think that it would be very feasible to use this in schools. It is usually very costly, and the sensitivity and – the sensitivity and specificity is not very good at present time.
7. What is the prevalence of dyslexia?
So we know that the prevalence of dyslexia is something — and that’s a little bit debated — but maybe between eight and 10 — eight to 12 percent. So if you have a classroom of 20 children, you will have two kids who will develop dyslexia in your classroom if you just look at the odds.
8. What are other reading challenges we should screen for?
I think what’s also really important that I don’t want to forget today is that we should not just identify children at risk for dyslexia. We want to make sure we identify children at risk for all reading impairments. That includes children who may struggle with reading fluency and reading comprehension — early reading — because of their environmental circumstances.
So they are coming maybe from families where there is not a lot of rich vocabulary, where the caregivers are speaking in very short sentences, and so you want to make sure you screen for these language components, as well, so that we can identify all children who are at risk for developing problems with reading, and not just the kids who have dyslexia.
How are the brains of people with dyslexia different?
There has been great research on the neuroscience of dyslexia showing brain characteristics in school-age children and individuals that have a diagnosis of dyslexia. A lot of the work we’ve done is to see whether these brain characteristics are a result of failing to learn to read or whether they predate the onset of reading. And our research — in addition to many other labs’ research — has shown that these brain characteristics are there before the start of kindergarten.
So the child has a — starts their first day of kindergarten — with their little backpacks — with a less optimal brain to learn to read. So we could show this on the group level. It’s not possible to do this on an individual level. So you cannot go to Boston Children’s Hospital and say, “I want to know whether my child Johnny will develop a reading disability or not.”
We’re way, way far away from this. I’m not sure we will ever be there.
What are the odds that a parent with dyslexia will pass it to their children?
So we always tell schoolteachers, administrators, that the absolute cheapest screener out there – it’s not the best screener, it’s not the screener we highly recommend — but if you have no budget, or let’s say your administration does not support screening – the cheapest, easiest thing you can do is ask every family in your classroom, “Did anyone in your family grow up – did anyone in your family struggle learning to read while growing up?”
And that will be a really important indication for you. If you’re a parent and you have an older child with dyslexia or you have dyslexia yourself, I highly recommend that you really pay close attention to the early reading development of the sibling. So are all the literacy milestone in place in preschool?
If you see that they’re not developing as their typical peers, I really would press really early to get evaluation, accommodations, et cetera, because it’s 40 to 60 percent of children who have an older sibling or parent who will develop dyslexia themselves. If the kid has two parents, the research has shown that the risk goes up to 75 percent.
What can we learn from the brains of infants about dyslexia?
So we have the BabyBOLD study, which is a study that is an extension of the BOLD study. So in the BOLD study we followed preschoolers over many, many years who either have a familial risk or not of dyslexia. And the question we ask there is, are the brain characteristics of dyslexia already there? Do they predate the onset of kindergarten or not?
In that study we could show that, yes, the brain characteristics of dyslexia could be detected as early as preschool. But the logical next question is, is this developing in parallel and in a relationship with language? Or is it something that the baby is born with? And so we are currently doing BabyBOLD, which tracks infants all the way from four to six months to late elementary school.
And we’re tracking them over time, and we look at their brain development. We look at their language and pre-reading skills. We have one — so one result that already came out of the BabyBOLD study is that infants who have an older sibling or parent with dyslexia already show differences in the white matter tracts in the brain compared to children who do not have a familial risk.
So the white matter tracts are almost like the highways in our brains. They connect two cortical areas like a highway and make it possible that information can flow from one area of the reading network to another area. And we could show that some of these tracts show altered — some — and we could show that some of these tracts show alterations in children who have a genetic risk compared to their peers, and these alterations are similar to older kids who have a clinical diagnosis of dyslexia.
So what we don’t know is whether these kids who have a genetic risk actually develop dyslexia later, right? So we have to see what is the sensitivity, specificity of this finding. Is it simply that everyone who has a familial risk shows these alterations, or is it specific to the ones who then will struggle to develop dyslexia?
So that’s why we have to track them longitudinally until we know whether they will be good readers or poor readers.
How can brain research help inform teaching?
I teach at the Harvard Grad School of Education, so I do a lot of teacher education myself. And usually the first day of our class on typical-atypical neurodevelopment for teachers, we ask the students, what do you want to learn in this class?
And a lot of the teachers say, “We want to learn all the brain-based tools so we can then bring them back into the district.” And our mission for the semester is to make them — well — our mission for the semester is to educate them that there are no brain-based tools, but that the brain can tell us a lot about development that is really valuable for their teaching.
But there is no brain-based tool. And so just learning how the brain develops, what is typical developing — what is a typical developing brain, what’s an atypical developing brain — gives you a really good foundation of — for your teaching, but it should not replace your teaching. It should not at this point directly inform curriculum.
It should just be more seen as a really good foundation. However, I don’t want to — I’m doing a lot of neuroimaging myself in the lab — I want to emphasize that we learn a lot from neuroimaging in terms of the underlying mechanisms of dyslexia, underlying mechanisms of how the reading brain is forming, et cetera.
So this is more telling the etiology — where dyslexia is coming from — which will inform how early we should screen, how early we should intervene, what kind of interventions may be helpful — these are some of the things that brain imaging could help with. But it will not be an easy bridge to make. But definitely a worthwhile bridge.
How have your ideas about dyslexia evolved?
So when I started — let’s say 15 years ago — I really thought that maybe imaging could help us to identify children who have dyslexia or identify children who will be at risk for dyslexia. I think we learned that it’s a lot more complicated than we thought – and that we have to do more studies longitudinally where we track children over time and move away from these snapshots where we just say, “This is how dyslexics do in third grade.” No, this is how one child with dyslexia in one group does in third grade. And I think we need to more look at developmental trajectories and have the whole timeline in mind.
Also I think we did not have protective mechanisms on the radar. Environmental variables such as — home literacy, how many books do the kids have at home, the quality of reading at home, not just the quantity of reading, neighborhood variables, socioeconomic status — all these have really strong influences on whether you are becoming a good reader or struggling reader.
And I think we have to start seeing the child as a whole and not just seeing it as, “This is reading,” and taking it out of context. And I think that’s something that I really learned in the last 15 years — that it’s a lot more complicated than we thought, and we have to take more variables into account.