Skip to main content

Mark Seidenberg

Dr. Mark Seidenberg is Vilas Research Professor and Donald O. Hebb Professor in the Department of Psychology at the University of Wisconsin-Madison. He is a cognitive scientist/neuroscientist/psycholinguist who has studied language, reading, and dyslexia for more than 30 years.

In addition to publishing more than 100 peer-reviewed articles in scientific journals, Dr. Seidenberg is the author of the 2017 book, Language at the Speed of Sight: How We Read, Why So Many Can’t, and What Can Be Done About It.

Dr. Seidenberg’s research explores basic questions about the nature of language and how it is acquired, used, and represented in the brain. One part concerns reading, how reading skill is acquired by children, and the causes of dyslexia. The second part concerns spoken language, particularly how it is acquired and the mechanisms underlying comprehension. He also studies the persistently low reading achievement of minority children, many of whom are from low-income backgrounds.

Learn more about Dr. Seidenberg’s research at the Language and Cognitive Neuroscience Lab.

Listen in to Why We Stopped Teaching Children How to Read. In this audio interview with John H. McWhorter for the Lexicon Valley series, Seidenberg discusses the fallout from the reading wars.


What has scientific research taught us about how children learn to read? And how well has that penetrated the classroom?

I think scientific research on reading and the issues that go along with it, how children develop, how they vary, the kinds of skills that reading draws upon has progressed enormously and has yielded findings that are absolutely rock solid.  So an example would be … we’ve known, studies of children’s behavior and adults’ behavior, and now more recently from studies of the brain, that the hallmark of skilled reading is the integration of print with what a person knows about the spoken language.

A child who’s learning to read already knows something about spoken language, quite a lot, and their problem is to figure out how print relates to spoken language. A child who is learning to read does not relearn language. They learn how print, this new code, relates to the language they already know. I think the neuro imaging research is really compelling.  Look at the brains of people who are better readers versus weaker readers. Or when you looked at developmental change from being a beginning reader to a skilled reader. The pattern that you see is that the people who are more skilled have integrated spoken and the written language at a neural level.

It almost, if you are a skilled reader it really doesn’t even make sense to talk about language and print as separate kinds of codes, because they are so deeply integrated throughout the brain. Well, if the integration of these systems is the hallmark of skilled reading and if educators have been paying attention to this research, would we have had reading wars over whether we should be encouraging methods that promote linking print and sound?  No, because we know that that’s actually a hallmark of what it means to be a skilled reader.

What we should be thinking about is what’s the fastest and most efficient way to get the most kids to get through this stage of integrated, getting to see how print relates so that they can get on with the task of reading for various purposes and learning from what they read. So there’s a case where the research developed over decades, got more advanced as the methods became more advanced. It’s not at the level where any reasonable person could debate. It doesn’t turn on one lab or one funding agency or any one method.

It’s an overwhelming success in the science and yet it’s had almost no effect on how educators think about reading, and indeed people are quite willing to keep fighting the reading wars endlessly.

Does reading work differently in different languages?

Reading works pretty much the same way despite differences in languages and learning systems. So there are a lot of different learning systems in the world. There are a lot of spoken languages that have very different properties. And, yet, if you look at how reading works in those different systems, there’s some minor variation, but essentially we’re talking about the same system.

I find that really remarkable, because when you look at the surface it seems like, oh, this writing systems are so different and the sounds of the languages are so different and the way they organize meaning is different.  And yet the underlying systems that support skilled reading are the same. That was a pretty amazing idea.

When children have problems learning to read, there are a number of causes, and they vary in severity and how malleable they are, how amenable they are to change or intervention. But there’s a short list of them. And, of course, these conditions can occur regardless of the language the child happens to be exposed to or the writing system they’re going to have to read.

The reading difficulties are caused by the same problems universally. There are some differences in the kinds of behaviors they effect. So, for example, in English the kinds of behaviors that are affected are children have difficulty learning about spelling and how particularly spelling relates to pronunciation because we have this writing system that has some high degree of irregularity. So in children with reading difficulties you’ll see lots of mispronunciation errors.

In writing system where the spelling and the pronunciation are much more systematic and have fewer exceptions, the problem isn’t learning how the spelling relates to the pronunciation. Children can learn those.  But it’s the speed of responding. The same underlying deficit is producing two different patterns of behavior.  It’s not different causes. It’s just that it interacts with the properties of the writing system and the language.

So you will see some surface differences in kind of symptomatic kind of behavior so to speak. But that’s pretty well understood.  And it’s coming from common underlying causes.

Does the reading brain of one person look like the reading brain of another?

As far as we know, the brain organization for reading is … there’s one. There aren’t multiple ways the brain can be structured to read. Now there are differences in skill. There are differences in how well developed these systems are. It’s not that everyone is identical. But we don’t find evidence that people’s brains can be organized in very different ways and yet still support skilled reading.

So, yes, there do seem to be certain universal properties of how reading works in the brain, but it’s not that the brains of the person reading Mandarin are different from the brains of the person who’s reading English.

This question of whether the same brain system underlies different writing systems is relevant also to the question of individual differences among children who are learning, for example, English. So educators just have a pretty firm belief that every kid is different, and that the challenge of teaching is to be able to tailor your instruction for the needs of the child and the learning style, the way they happen to learn best.

We don’t find that when we study the computational and behavioral and neuro systems that underlie reading. We find that there’s a lot more commonality then there are differences. There are differences in skill. There are differences in how much experience kids have had. There are differences in how helpful their instruction has been.

But we don’t find that the brains of the children are organized in highly idiosyncratic ways that mean that mean that one child will learn to read one way and another child will learn to read another way. There, it could have been that way. That’s how the research could have turned out, but it just hasn’t.

What are schools getting wrong about teaching reading?

There’s a belief that children will learn to read if they are placed in an environment in which they are motivated to read, in which reading is highly valued, in which there is time for reading, in which there are engaging activities around reading, much like learning to, the way that the child would learn his first language and that people don’t, on the education side, don’t think they’re in the business of teaching reading.

Teachers, prospective teachers are not taught methods for teaching reading. That is not part of their training. Children are left to absorb reading in a reading, literacy rich environment.  And teaching reading is sort of thought of as like the wrong thing to do, because that’s just focusing on the mechanics. And on the education side most of the attention is focused on comprehension, like how can we encourage kids to read, understand it, question the material that they’ve read.

On the reading side the evidence says getting the child into the system is really a hurdle for many kids, and that is where effective instruction and guidance and feedback is really important. That’s exactly the wrong time to be leaving the child to discover how the system works. We now want to discover how writing systems work. You know it took so long for writing systems to be invented and to evolve into what they are now … we don’t want to place the child in the position of having to figure out how this system works.

That’s why the instruction is really important. And that’s exactly what the kids are not getting. Teachers aren’t taught to do it, and they aren’t encouraged to do it. And it’s not thought to be necessary because the child will figure this out in the appropriate environment. And that turns out not to be right in a fundamental way.

Comprehension on the other hand, if the child can figure out how print works and relate it to things they already know about spoken language, they get comprehension at the beginning for free. They already know how to comprehend language. The problem is they don’t know how the letters works. They don’t know how the print works. There is a lot to learn there. That’s something where we could be efficient and effective in helping the child to get over this hurdle, and that’s where skilled instruction would really make a difference.

So it’s just deeply unfortunate that teachers are kind of socialized into the belief that I should just stand back because a kid’s going to converge on this as they love, they want to be literate. And for many kids it takes more than that.  Most kids.

How do we help teachers succeed in teaching reading?

It’s very hard to make changes, and I think we have dug ourselves as a society a deep hole here. I can identify the problem. And maybe that’s helpful. But educators know that they don’t have to pay attention to certain kinds of research because it’s not relevant to what they do. That’s the leap. But that means that they don’t know what they don’t know. They don’t know, they are closed off from a body of work that would be helpful, that would make it possible for more teachers to succeed and the kids to succeed.

I think there is a problem with socializing people into this sort of view that teaching is an art and that you can only learn on the job. Learning how a kid reads from just observing them and monitoring their behavior is a really tough way to learn how to be an effective teacher. So all I can say is that it’s not a competition for the child’s brain.

We all want them to succeed. And it’s shameful to me that there’s this … it’s sad, it’s pathetic that there is such a vast body of research, which could be translated into practice. It doesn’t come with educational implications attached to it.  It’s something that requires further work and cooperation. But it’s currently just being left on the table because of these beliefs that really have been promulgated generation after generation about how the scientists are really just, you know, they put their lab coats on, they’re measuring things, and they really can’t capture what we do as teachers.

I think that’s a fundamental error, and if we could only get past our history, we could be doing what we all want, which is to have more kids succeed.

How does a teacher’s expertise contribute to helping kids read?

Teachers are experts about what happens in the classrooms. Teachers have far more experience interacting with the children many hours a day. Of course they have their own kinds of expertise that scientists, most scientists wholly lack.

There are some great science that’s being done by people who are former teachers of course. But in general teachers know much more about, I think, the issues related to motivation, attention, what kids will really go for. They have an enormous amount of expertise that could be, that would be essential to bring, to integrate with the science. It’s not that the science comes with the recommendations attached.  There’s additional steps, several steps.

And the teachers’ expertise is invaluable. It’s just that, just being a teacher and observing children isn’t really adequate as a way of figuring out how kids learn to read and could lead you to some very beliefs that just aren’t really advantageous for the kids.  So they have enormous expertise that the scientists lack. And the idea, of course, is to bring these groups together, and the surprising thing is how incredibly difficult it is to do.

Is there an agreed upon definition of dyslexia? What are the key features of dyslexia?

The issue of definitions of dyslexia is one that people really care a lot … I have to say definitions of complex things are kind of overrated. So I can give you a definition of dyslexia. Dyslexia is a developmental condition that interferes with children’s ability to learn to read. It’s got genetic and neurobiological bases. And it can also often be, a child’s problems can often be ameliorated with timely and appropriate interventions.

That doesn’t tell you anything about what causes the condition, how variable the kids are, any of the important stuff. It’s like trying to define language. We can define language, this communicative system that only humans use.  But that doesn’t tell you anything about the important properties of language. So people do have to define dyslexia for various purposes, like for the purpose of paying for medical care or deciding how to allocate educational resources and other things.

Definitions have their place. But the fact that it’s a difficult condition to define in a few words doesn’t mean that it doesn’t exist or that people are unclear about it. It’s a complex condition. There’s multiple causes.  It interferes with reading. Sometimes depending on the severity of the condition, it can also interfere with children’s learning of other things.

How do early language experiences contribute to a child’s ability to read?

Children’s learning to read is hugely dependent on experiences that they have before reading instruction, school, or any kind of school experience related to reading starts. Those experiences that the child has with spoken language and with the world that that language refers to are enormously important. I, myself am extremely concerned about and focusing more and more my attention on the differences that we see in children before they even get to school with respect to spoken language.  Things like vocabulary and when you know a word, how much you know about the word, sentence structures, and other things.

Children are behind with respect to spoken language development or at least disadvantaged with respect to spoken language development and, therefore, their readiness for reading on the first day of school.  So it’s absolutely essential that we pay attention to children’s language development, and that of course means also their development of the knowledge that the language conveys and how it relates to the world.

There are these huge individual differences in children in terms of this early learning, and their consequences for reading are clear. They are hard to overcome. They are hard to overcome. There are huge vocabulary differences, and by the time the child gets to school without enough time to, there isn’t enough time in a classroom to teach the kids all the words they don’t know.

So once a kid falls behind in these areas, it’s actually very challenging to help them catch up. That’s where my own research is really focused right now. So all the attention that’s being given, I think, in the public consciousness about the importance of talking with your child, the importance of reading to your children, reading with your children, interacting over books and so on is absolutely spot on and is really critically relevant to how they’re going to do with reading and other aspects of school.

One of the amazing research findings from just recent history is we all thought for a long time that reading to children is important, which it is. Then we thought it introduces children to reading in this pleasurable way and it’s a great experience for the caregiver and the child and it introduces them to stories and they begin to understand how books work and all that.

It does that, but it does a lot more. So if you start, people, researchers, psychologists have started to look at the language of children’s books. The language of children’s books is very different from the language of speech.  The first line of any children’s books … no. What’s the first line of the Madeline book?  “An old house in Paris that was covered with vines where 12 little girls in two straight lines” … people don’t talk like that.

When you systematically actually look at children’s books, they have words, concepts, sentence structures, talk about things in the world that children are not going to get from their immediate experience. So one of the added benefits of reading to your child is a kind of language enrichment.

You are teaching them words like, well, “Eiffel Tower” comes to mind. But, you know, the salami in The Very Hungry Caterpillar or, you know, the vocabulary that’s used in children’s books is actually not the same as we use in every day speech. And so when you are reading to your child aloud, you are giving them a kind of speech they’ll learn from that they wouldn’t otherwise get.

This is an important vehicle for developing their language and their knowledge of the world. And it’s also one that we potentially make better use of to try to close some, prevent some of these language-related gaps.  Maybe children’s books could be a really terrific vehicle for doing that.

What are the differences between reading and speech?

I think people don’t fully appreciate the extent to which reading printed code and speech are similar but different. So they’re similar because they’re both ways of representing a language, but they’re not actually identical.  People don’t talk the way books are written, the way we write. So in the beginning most of what children know about language comes from speech.  That’s what they learn first.  And their problem is to get reading to catch up.

But then later on, because reading and speech differ, the main vehicle for learning more about language is reading. You’re going to encounter much more complicated sentence structures and words and expressions and so on.  So there’s this interesting, there are these interesting tradeoffs between what we read and what we hear or say that change over time.

I think most people just think of them as just different ways of doing the same thing. But that’s not exactly true. And that’s why it’s so important for people to understand the relationships between the two, to have it kind of deeply imprinted in their brains.

How do a child’s experiential differences impact their reading?

I do think that experiential differences that children have are really crucially important. You know, I got some email from a teacher who said, “I don’t understand what’s happening. My children, students have learned to read. I’ve tried to do the right things. They can sound out words. They can pronounce lots of words correctly. They seem to understand them. But then when they read books, they don’t comprehend what they’re reading.”

My thought is, how can this be?  If they understand spoken language and they can figure out how it relates to spoken language, then why couldn’t they understand the text?  After a little bit of back and forth, it turned out the answer was, they didn’t understand the topic. They didn’t know anything about what the story was about or the text was about.

So I think it’s, one thing that I really have come to appreciate much more is the extent to which, yep, you need to learn how the mechanics of reading work, you need to have reasonable spoken language skills, and you also need to know something about the world that we’re using this language for.

The other thing I think I’ve come to appreciate much more, and this is something that wouldn’t be news to teachers, is that motivation is just enormously important. So we understand how reading works to a pretty high level. We could think of ways to incorporate that into the classroom practices, but there’s a kid there. And that kid has to be motivated, and that kid has to be engaged.

That kid has to know why they’re doing it and what it’s for. And I do think sometimes as scientists we have not paid enough attention to that.  It’s crucial.

How can we bridge the divide between the science of reading and what happens in classrooms?

I’m stumped in the way about how to move forward, in the way that we’re stumped about how to move forward on climate change or on any other controversial issue about which people really, really have very different beliefs. 

I like to think that people have the same goals and want the same things. And perhaps if we find small ways to do better than we have, that will open the door to the much bigger changes that we really need.  I’m hopeful that we can still do that.

On the education side you need to make it easier for teachers to succeed. And we need to do more than pay lip services to the value of the profession. We say that teachers are essential and teachers are so important to our children’s development, but then we don’t actually support them or pay them or do other things that make the job actually attractive.

In some cases they’re demonized as part-time workers who … and it’s thought that anyone could teach, which isn’t true. And so I would value teaching at every level. Not just who goes into the field and how they’re trained, but also support that they get on the job to do what is a difficult job.

I’m committed to the idea that knowledge is good and that, indeed, what we’ve learned about children and development and language and reading and math has, it’s one of the real great success stories in science. And so surely we can make use of this information to make the world a better place. I really do believe that we can do that.

And we have done that in other areas. Medicine, for example. But it’s going to take some taking down of some really high barriers and some real energy and effort, and I hope we can succeed.

How would you change the way teachers are prepared?

I would go back to the idea that teaching is a skill, a very sophisticated skill, that it requires special training and that the professional training includes things like relevant background in things about child development, motivation, language, how reading works and so on.

I would go back to the idea that just like in other professions there’s a certain body of knowledge that teachers need to know as preparation for going out to do this difficult job. I would abandon the idea that teachers can’t be taught to teach, which is so prevalent in schools of education. Of course they can be taught to teach, and then they’ll refine that when they get to the classroom.

Right now what I hear from teachers is about how irrelevant most of their preparation was and how lost they felt when they got there. It’s not just anecdotal. Of course there are studies that show this. So I would go back to the idea that, let’s respect the profession and let’s introduce a curriculum that actually will be, includes the relevant sorts of stuff.

It would include some of the things teachers are already being exposed to, like, for example, the fact that the kids in their classrooms may come from a wide variety of backgrounds, have experiences that are, life experiences that are quite different from the teacher’s own that are going to influence how that kid functions in school. That stuff is essential, too. It’s just not sufficient.

So I would just make teacher training serious, teacher education a serious enterprise, like other kinds of professional training.  And to the people who say it’s just something learned on the job, that’s just not okay.  That’s like saying learning to drive is just something you do by doing it, or, you know, becoming a doctor, well, you just learn that on the job.

Let’s say, recognize that teaching is also a skill that requires elaborate sorts of, an elaborate set of abilities and that people need some preparation for a job.  And, of course, we can provide it, because we know quite a lot that’s relevant. That’s something we could do. That’s something we can do tomorrow.

What are the challenges of learning to read in a different dialect?

What we are aware of is that the language in the home and the language in the school can differ. People have known that for a long time. A child speaks Spanish or Mandarin or French in the home, and then they go to school in English. Those things are different.

And so, you know, bilingualism is this great thing in general, but, you know, a child who’s been speaking one language and then has to go to school in another language faces additional challenges.  And so when they fall behind, there’s going to be a reason: they’re still learning English. The same thing happens though with dialect variation.

So there are dialects spoken in different parts of the country, and then there’s the standard dialect of the school and in particular of the books. Books are written in a sort of standard normative form of English, a particular dialect of English. It’s called Standard English. But it’s just the one that we’ve decided to use.

Children from various backgrounds may speak dialects that are different from that in many ways. They’re still English, but they … words are pronounced differently, different words are used, the syntax is different.  The ways language is used are different. And so a child who’s coming from that kind of background, where they’re speaking a different dialect and then has to go to school and learn the school dialect — just like a bilingual — kid has that extra burden. And I don’t think that’s recognized.

So it is one of the factors that, for example, arises with African American kids where African American English is a dialect of English. Linguistically, it’s like other dialects in other languages. It’s just a variant of English.  But it happens to differ from the one that’s going to be required for school. Those children can learn both dialects; but it’s going to take more time, and it’s going to take more effort.  They have more to learn.

Unlike a kid who speaks Spanish in the home who has more to learn to get up to speed in school English, the child who’s speaking a minority or non-standard dialect in the home also has more to learn, but that’s not actually recognized.  So those kids are at a disadvantage for things that have nothing to do with, they’re separate from poverty or from culture or from the individual abilities of the child. They have to do with the circumstances of the language of the home differing from the language of the school.

I think we can do much more to recognize that those kids are being compared to kids for whom the dialects are the same and that’s not a fair comparison.  And moreover some parts of this achievement gap are built into this kind of set of circumstances. It also means we could do something about it. Children can get exposed to the full range of dialects of English before they get to school, not be quite so stunned by the differences between what the way they talk and the way the teacher talks. And I think that also would be helpful for reading and school success in general.

Books by this author

Find this author’s books on these booklists