Language at the Speed of Sight

How We Read, Why So Many Can't, and What Can Be Done About It


By Mark Seidenberg

Formats and Prices




$17.99 CAD



  1. ebook $13.99 $17.99 CAD
  2. Trade Paperback $19.99 $24.99 CAD

This item is a preorder. Your payment method will be charged immediately, and the product is expected to ship on or around January 3, 2017. This date is subject to change due to shipping delays beyond our control.

In this "important and alarming" (New York Times) book, see why so many American students are falling behind in their reading skills while others around the world excel.

The way we teach reading is not working, and it cannot continue. We have largely abandoned phones-based reading instruction, despite research that supports its importance for word recognition. Rather than treating Black English as a valid dialect and recognizing that speaking one dialect can impact the ability to learn to read in another, teachers simply dismiss it as "incorrect English." And while we press children to develop large vocabularies because we think being a good reader means knowing more words, studies have found that a large vocabulary is only an indication of better pattern recognition. Understanding the science of reading is more important than ever–for us, and for our children. Seidenberg helps us do so by drawing on cutting-edge research in machine learning, linguistics, and early childhood development. Language at the Speed of Sight offers an erudite and scathing examination of this most human of activities, and concrete proposals for how our society can produce better readers.



       CHAPTER 1

     The Problem and the Paradox

I DON’T THINK WE’VE MET, but I know two things about you. One is that you are reading these words. And because you’re reading these words, I also know that you are an expert at what this book is about. Which is reading.

A friend once gave me a scholarly 380-page book about the history of the pencil. I like pencils. I’m good at using them. I like a fresh pink eraser tip. I just do not need to read about how they got that way. Perhaps reading is for you like the history of pencils is for me. Why read about reading?

Reading is one of the few activities you do every day whether you want to or not. Street signs, menus, e-mails, Facebook posts, novels, ingredients in Chex Mix. You read for work, for school, for pleasure; because you have to, because you want to, because you can’t help it. That is a lot of practice over a long period. If it takes thousands of hours to become an expert at something like chess, we readers are in grandmaster territory.

One result is that everyone has an opinion about how they read. When you are introduced at the dinner table as someone who studies reading for a living, you hear all the stories. People tell me whether they read word by word or in big chunks. Whether they read visually or hear the sounds of words in their heads while they read. How they learned to read and how their kids learned. Many confessions from people who say they are slow readers, more often than not from accomplished people in fields that seemingly demand a high level of reading competence—lawyers or school superintendents, say. Most of you think that other people read faster than you. It’s the opposite of Lake Wobegon: reading is the land where the folk are all below average.

But here’s the rub: people manage to be good at reading without knowing much about how they do it. Most of what goes on in reading is subconscious: we are aware of the result of having read something—that we understood it, that we found it funny, that it conveyed a fact, idea, or feeling—not the mental and neural operations that produced that outcome. People are unreliable narrators of their own cognitive lives. Trying to understand reading by observing our own reading is hopeless, like trying to understand how a television works by watching Game of Thrones. Being an expert reader doesn’t make you an expert about reading. That is why there is a science of reading: to understand this complex skill at levels that intuition cannot easily penetrate.

I am a psychologist/psycholinguist/cognitive neuroscientist who has been studying reading since the disco era. I’m not alone: a huge community of scientists studies reading around the world. Many people are surprised to learn there is a science of reading. Really, what is there to study? Words on page; eyes scan words; words are comprehended. It’s just like listening, only visual. Book ends here. Beneath this seemingly simple behavior, however, a vast, coordinated network of activities is occurring. A snapshot of a person’s brain activity while reading shows that most of the brain is involved: areas involved in vision and language, of course, but also neural systems that control action, emotion, and decision making; several memory systems; and much of the rest. Forget the myth that people only use 10 percent of their brains: we use more than that merely reading. The relationship between the experience of reading and its underlying neurocognitive mechanisms is about as opaque as the relationship between behavior and its psychodynamic causes in Freudian theory. Fortunately we have better methods than Freud for exploring the subconscious basis of behavior. We’ll have you lie down and talk to us, but in the barrel of a magnetic resonance imaging machine, not on a couch.

We’ve learned quite a lot actually.

We understand the basic mechanisms that support skilled reading, how reading skill is acquired, and the main causes of reading impairments.

We know which behaviors of three- and four-year-olds predict later reading ability.

We know how children become readers during the first years of schooling and the obstacles that many encounter.

We know what distinguishes good from poor readers, younger from older skilled readers, and typical readers from those who are atypical because of constitutional factors (such as a hearing or learning impairment) or environmental ones (such as inadequate instruction or poverty).

We understand what is universal about reading (things that all readers do the same way because their brains are essentially alike) and what is not (because writing systems differ).

We have identified the main neural circuits involved in reading and some of the anomalous ways they develop in children with reading impairments.

We even have computational models of learning to read, skilled reading, dyslexic reading, and the loss of reading ability due to brain injury. It takes a deep understanding of these phenomena to develop models that reproduce them.

This vast research base has led to the development of methods that can reliably help many children who struggle to read. Researchers disagree about many details—it’s science, not the Ten Commandments—but there is remarkable consensus about the basic theory of how reading works and the causes of reading successes and failures.

Sputnik Lands on USA

The successes of reading science create a paradox: if we know so much about reading, why are literacy levels in the United States so low?

In 2011 America had what our president called a Sputnik moment, occasioned by the release of the latest results from the Programme for International Student Assessment (PISA), a massive appraisal of the reading abilities of fifteen-year-olds in seventy-four countries and municipalities from Kyrgyzstan to Canada. As in previous rounds and again in 2012, US performance was close to the average for the thirty-four member countries of the Organisation for Economic Co-operation and Development (OECD), which conducts the PISA exercise. However, the 2009 round was the first to include data from Shanghai and Singapore, which scored higher than the US, as did Asian neighbors Korea, Hong Kong, and Japan. The Shanghai students lapped everyone else, scoring highest in reading, math, and science by wide margins. These results received far more attention than the fact that the US always scores lower than countries like Australia, Canada, Finland, and New Zealand in such exercises. Government officials and commentators treated the results as a wake-up call about the uncompetitive state of American education, said to pose a threat to the country’s future akin to that represented by Sputnik in 1957. The Soviet Union’s stunning success spurred a rapid, comprehensive governmental response that seems as much a relic of a bygone era as the satellite itself. Within two years, legislators had created the National Aeronautics and Space Administration and the precursor of the Defense Advanced Research Projects Agency (which funded the development of breakthrough technologies including the Internet), tripled the National Science Foundation budget, and allocated hundreds of millions of dollars for student loans and scholarships under the National Defense Education Act. But 2011 was not 1957, and the second Sputnik moment passed quickly, rapidly dropping out of public discourse (Figure 1.1).

Although the PISA results made the news, there is plenty of in-house data about America’s literacy issues. The country is a chronic underachiever. A 2003 study found that about 93 million adults read at basic or below basic levels. At those levels, a person might be able to follow the instructions for mixing a batch of cake mix but not understand a fact sheet about high blood pressure. The emergence of our literacy problem is visible in the performance of fourth and eighth graders on the National Assessment of Educational Progress (NAEP), “the Nation’s Report Card,” an assessment administered by the US Department of Education. Over half the children have scored at basic or below basic levels every time it has been administered. At the upper end, we are turning out fewer highly proficient readers than expected given our economic resources. Like everything else about education in the US, the results of assessments like the PISA and NAEP have generated controversy. Are the tests too hard? Are they poorly matched to what American children are taught? If the scores are so low, why does the US continue to have the biggest national economy in the world? The American polity has been in a test-happy phase, and much other data about who can read and how well paints a consistent picture: large numbers of individuals in the US read poorly, as has been true for many years. Although I focus here on the US because it is home, the situation is similar in many other countries with advanced economies.

FIGURE 1.1. The number of times the phrase “Sputnik moment” was uttered on CNN, the American news network, over a two-year period. The large spike followed the release of the 2009 PISA results and coincided with the president’s State of the Union address.

The consequences of marginal literacy for the affected individuals and for society are vast, as we all know. Reading is fundamental—so the slogan goes—and children’s failures to acquire reading skills have rapidly cascading effects on learning other subjects. Even math is implicated, given the heavy emphasis on language in math curricula and instruction (word problems, explaining your work, the practice of embedding math problems in real-world contexts). Having less reading experience makes it harder to learn how to learn and how to think critically and analytically. Reading failures place children at high risk of falling out of the educational system. As adults, poor readers cannot participate fully in the workforce, adequately manage their own health care, or do much to advance their own children’s education. If that doesn’t convey sufficient urgency, a 2012 report from the Council on Foreign Relations declared, “Large, undereducated swaths of the population damage the ability of the United States to physically defend itself, protect its secure information, conduct diplomacy, and grow its economy.”

This information plagues me, especially because the situation is not new. Scores on standardized reading tests have been nearly flat for decades. Hopes that the Harry Potter craze would increase children’s involvement in reading gave way to deep disappointment when it didn’t. We hear the suspicion voiced everywhere—from Ivy League colleges to local middle schools—that important reading skills and habits not measured by standardized tests, such as the ability to engage in close reading of challenging texts, are in active decline. According to Philip Roth, that game is already over. “The evidence is everywhere that the literary era has come to an end. . . . The evidence is the culture, the evidence is the society, the evidence is the screen, the progression from the movie screen to the television screen to the computer. . . . Literature takes a habit of mind that has disappeared.”

Anxiety about reading achievement underlies endless debates about how reading should be taught. Parents know it is essential that their children learn to read well, but lacking confidence that schools can do the job, they are driven to seek additional help from tutors or commercial learning centers, if they can afford it. Our knowledge about reading has grown enormously, but, as ever, many people cannot read, or can read only poorly, or are able to read but avoid it. And so I have asked myself whether our science has anything to contribute to improving literacy outcomes in this country and elsewhere.

It might not. Reading failures could simply be collateral damage caused by other, deeper problems, such as poverty. About 15 percent of the US population lives in poverty, as the US Census Bureau defines it, including over 16 million children. Poverty is associated with higher infant mortality, higher risk of atypical neurodevelopment, shorter life span, worse health and health care, higher crime and incarceration rates, lower educational achievement, higher dropout rates, poorer schools with weaker teachers, and, at the bottom of a much longer list, poor reading. Surely addressing poverty would have a bigger impact on literacy than anything inspired by our research.

Poverty is a huge factor in America, as in other countries. Reading achievement is higher in Australia, Canada, and New Zealand than in the US, but not among their aboriginal peoples, for whom poverty rates are high. No discussion of what to do about reading can ignore the noxious conditions that govern many people’s lives. However, 93 million is a lot of marginal readers, and they are not confined to the lowest economic stratum. There’s Ron Gronkowski, a colorful professional football player, who said he hadn’t read a book since Mockingbird to Remember in ninth grade. But look around: they walk among us. We could be talking about an airport security agent, an X-ray technician, police officers. High school graduates who cannot handle the literacy demands of community college. “Student athletes” who play big-time college football. “Low-information voters” whose primary news sources are talk radio and cable news. Or perhaps some readers of this book, highly motivated yet struggling through it. People with only basic reading skills can hold demanding jobs that affect health care, personal safety, law enforcement, business, politics, and every other element of society.

How much of the problem is due to cultural characteristics that conspire against our becoming readers, such as the profusion of screen-based activities: 24/7 movies and television, gaming, apps, cat and music videos, PewDiePie (45 million YouTube subscribers, 2,900 videos)? Are reading and writing gradually being reduced to tools in the service of reading-ish activities such as texting, tweeting, and Facebooking? What about overscheduling: the relentless demands of ballet, piano, violin, drama, tae kwon do, 6 a.m. slots at the hockey rink, soccer practice, matches, and travel? Middle schoolers cannot read while running dribbling and passing drills. Perhaps reading is simply less important than it used to be. Writing was the first information technology, but now there are others. We have screens, they have pictures, sound, video. Text carries less of the communicative load than before. I greatly prefer to read a recipe, but a video demonstration conveys additional information, for people who have the patience to sit through it. You can read about the phenomena I describe in this book but also see some of them in videos or try them out, which I highly recommend (and provide links to allow).

Something is happening here, but we don’t know what it is, yet. If you position your ear close to the Internet, however, you can hear the steady thrum of reports and pronouncements about the state of reading: the good (easier access to a larger quantity and greater variety of texts than ever, integration of text with sound and image), the bad (easier access to a larger quantity and greater variety of nontextual media and undemanding text), and the ugly (a long-term decline in the linguistic complexity of K–12 schoolbooks). The National Endowment for the Arts issues grim warnings about the decline of reading, while educators view reading as one of several “literacies” for the modern era.

The results of this real-world experiment in how information is transmitted will not be known for generations. I think it is safe to assume that reading is not going away any time soon, if ever, and that people who are skilled readers will continue to have advantages over people who are not. With this in mind, I focus on a factor that has direct, lasting effects on who reads and how well: education. With some exceptions, children learn to read in school. The only certain way to obviate low literacy is prevention: successfully teaching children to read in the first place. Would more people be better readers if they had been taught differently? How much does schooling affect how well children read and, with it, their engagement in reading?

Here we encounter a problem. There is a profound disconnection between the science of reading and educational practice. Very little of what we’ve learned about reading as scientists has had any impact on what happens in schools because the cultures of science and education are so different. These cross-cultural differences, like many others, are difficult to bridge.

The gulf between science and education has been harmful. A look at the science reveals that the methods commonly used to teach children are inconsistent with basic facts about human cognition and development and so make learning to read more difficult than it should be. They inadvertently place many children at risk for reading failure. They discriminate against poorer children. They discourage children who could have become more successful readers. Many children who manage to learn to read under these conditions wind up disinterested in the activity. In short, what happens in classrooms isn’t adequate for many children, and this shows in the quality of this country’s literacy achievement. Reading is under pressure for other reasons, but educational theories and practices may accelerate its marginalization.

Reading science cannot address all of the issues that affect reading achievement—all those sequelae of poverty, for example—but it does speak to serious issues about how reading is approached in schools and at home. A look at the basic science suggests specific ways to promote reading success. These do not require more testing or new federal laws; they do not require vast infusions of money; they are not based on classroom computers that treat learning like a video game or other faddish uses of technology. What they require is changing the culture of education from one based on beliefs to one based on facts.

The next several chapters provide an insider’s guide to some beautiful science that looks at reading at levels ranging from the movement of one’s eyes across the page to the brain circuits that support reading to genes that affect how those circuits develop. Some of it is paper-clip-and-string research using the simplest of methods, such as recording the errors that children make when they read aloud or having the subjects in an experiment decide if SUTE is a word (it isn’t, but it sounds like one, which creates momentary uncertainty even when it is read silently). Some of the research involves exotic methods such as transcranial magnetic stimulation, which briefly disrupts neural processing in parts of the brain relevant to reading. We will look at children who struggle to read and children who can read words but comprehend very little. Of course there is the terrible fascination of patients whose brain injuries impair reading in unusual ways. The late Oliver Sacks’s famous agnosic patient mistook his wife for a hat; a patient with an analogous reading impairment, called deep dyslexia, read the word SYMPATHY as “orchestra.” Along the way I cheerfully destroy a few myths about reading, including ones that have supported years of commercial schemes said to improve your child’s reading in a few short weeks or to make you a super-mega-reader who can whip through books a page at a glance. We will look at educational neuroscience, the study of the brain bases of skills such as reading and math and how they are acquired. Fact: in neuroimaging studies, poor readers show atypically low activation in a part of the brain that processes the spellings of words. Can such findings inform how reading is taught, or do they merely provide neural window dressing for something teachers already know, that poor readers don’t read or spell words very well?

With an understanding of the basic science of reading in hand, I take a close look at what happens in classrooms in America. If you are curious about the state of reading, these findings will, at the least, surprise you. If you are (or were) a parent of a school-age child, you might feel the shock of recognition and the dropping of scales from your eyes.

Understanding how the science relates to educational practices is an important step, but it is not enough to generate change. Educators are deeply immersed in their own worldview and well defended against incursions from outside. The education side focuses on “literacy” (literacy practices, cultural attitudes toward literacy, multiple types of literacy including ones that do not involve print), not reading. The scientific perspective is seen as sterile and reductive, incapable of capturing the ineffable character of the learning moment or the chemistry of a successful classroom. That people who enter the field of education do not gain exposure to modern research in cognition, child development, and cognitive neuroscience deprives them of the ability to evaluate what is studied, how it is studied, what is found, and what it means. Instead, they are exposed to the views of a few authorities, the most influential being Lev Vygotsky, who lived in the Soviet Union, wrote in Russian, died in 1934, and never saw an American classroom or a television, calculator, computer, video game, or smartphone. We scientists naively take the importance of our findings and their implications for classroom practice to be obvious, overestimating how much people on the education side know about or value them.

Education as a discipline has placed much higher value on observation and hands-on experience, which brings us back to page one (of this book). Intuitions about reading do not penetrate very far, and observations are influenced by prior beliefs, including what one has been taught about how children learn and how reading works. Educators are people too, subject to the same cognitive limitations and biases as everyone else. The lack of scientific literacy, combined with deep faith in the validity of personal observation, creates vulnerability to claims that are intuitively appealing but unproved or untrue.

People complain about the state of education about as much as they complain about the government. There’s an endless, hopeful, desperate search for a game-changing innovation that will make more people smarter, more quickly, with less effort and expense. Some people focus on eliminating poverty itself, which would obviously have enormous impact on education and much else. We can focus at the same time on how to modify more tractable conditions to produce better outcomes. Learning more about the values and beliefs of the educators who teach teachers, design curricula, and create instructional practices could be a powerful impetus to change. In the closing chapters I spell out these concerns as they have arisen in reading education. This analysis clarifies what needs to change, and I discuss some ways we could move forward.

Since change can’t be achieved overnight, I’ve also written this book to give readers—a parent interacting with a teacher, a school principal, or the bureaucrats who run the local school system; a voter whose representatives have the power to affect educational goals and practices, how public education gets funded, and how tax dollars are spent—information they can use now. These are the tools you need: an explanation of what we know about skilled reading, learning to read, the causes of reading difficulties, and the brain bases of normal and impaired reading; a critical analysis of the ways that reading is taught; an account of the controversies in reading education; and an understanding of educational ideology and how it evolved. Like health care, education is a multi-billion-dollar industry involving multiple stakeholders—government, business, educators, parents, children, taxpayers, unions, interest groups, philanthropy—whose perspectives and interests often conflict. In education, as in health care, end users benefit if they are informed, proactive participants rather than passive recipients.

Of course plenty of information about reading is available online, but on the Internet nobody knows your website is a dog. Websites do not come labeled with independent scientific seals of approval; doctrinal biases are not labeled as such; a layperson has no way to evaluate contradictory claims and assertions. Everyone—ranging from the National Council of Teachers of English (not much science there) to the Scientologists (whose teaching method L. Ron Hubbard is said to have devised himself)—has an opinion about reading. School districts post jargon-laden statements of teaching philosophies and curricular goals on their websites. It’s a jungle out there, and a person could use some help.

A Science of Reading

Reading is one of the oldest topics in experimental psychology—the first American professor of psychology, James McKeen Cattell, studied it in the 1890s—and yet it is also a favorite topic of researchers in cognitive neuroscience, a field that emerged about a hundred years later. Why this enduring interest in such a familiar activity?

Reading is unique. Reading is among the highest expressions of human intelligence. Although spoken language is usually taken as the capacity that distinguishes people from other species, researchers have debated the degree to which other species’s communication systems resemble language. No other species has a linguistic capacity equal to ours, but animal communication systems share some properties with human language. The late African Grey parrot Alex clearly had communicative interactions with his longtime trainer Irene Pepperberg. Was his use of words very much like human speech or an oddly evocative simulation, the result of thousands of hours of intense training? Whatever the answer to that question, we know that no other species has an ability remotely like reading. Indeed, Homo sapiens didn’t either until the invention of writing about five thousand years ago. Understanding this complex skill means understanding something essential about being human.

Reading is important. Human culture has evolved to the point where this skill is critical to our ability to thrive. For most of human history people were illiterate and yet functioned well enough. That is not the case in modern societies. In a 2007 article bemoaning Minnesota children’s low levels of reading achievement, Garrison Keillor observed, “Teaching children to read is a fundamental moral obligation of the society.” Given


  • "In Language at the Speed of Sight, [Seidenberg] develops a careful argument, backed by decades of research, to show that the only responsible way to teach children to read well is to build up their abilities to connect reading with speech and then to amplify these connections through practice, developing skillful behavioral patterns hand in hand with the neurological networks that undergird them...Every teacher of young children as well as those who train them should read this book."—Wall Street Journal
  • "Seidenberg...unravels the science of reading with great flair. He is the ideal guide - and it turns out that we need a guide to reading, even though we've been doing it most of our lives."—Washington Post
  • "Seidenberg reviews the latest science on reading and makes an impassioned plea for putting this knowledge to use."—Scientific American
  • "Cognitive neuroscientist Seidenberg digs deep into the science of reading to reveal the ways human beings learn how to read and process language.... Seidenberg's analysis is backed up by numerous studies and table of data. His approach is pragmatic, myth-destroying and rooted in science--and his writing makes for powerful reading."—Publishers Weekly
  • "The neuroscience underlying [Seidenberg's] findings is complex, of course, but [he] does not often fall into thickets of technicality...his discussions are clear and accessible.... A worthy primer on the science of comprehending language."—Kirkus
  • "No technologically advanced society exists without reading. This is the remarkable story of why and how it all works. From David Letterman's irony to posited Sumerian patent trolls, the writing is lively, informative, and supremely entertaining."—Daniel J. Levitin, best-selling author of This Is Your Brain on Music and The Organized Mind
  • "Have you picked up the idea that reading is something that kids 'just pick up' and shouldn't be rushed into it, or that learning to read is something different from mastering something separate called 'comprehension,' or that a whole book about reading would be dull? Language at the Speed of Sight will disabuse you of all three notions and more-pick it up and marvel at how hard it will be to put it down."—John McWhorter, New York Times best-selling author of Nine Nasty Words, Word on the Move, and Talking Back, Talking Black
  • "Few works of science ever achieve Italo Calvino's six qualities of our best writing: Lightness, exactitude, visibility, quickness, multiplicity, and consistency. Mark Seidenberg's new book achieves just that. If every educator, parent, and policy maker would read and heed the content of this book, the rates of functional illiteracy, with all their destructive sequelae, would be significantly reduced."—Maryanne Wolf, author of Proust and the Squid
  • "A world-renowned expert explains the science of reading with clarity and wit--anyone who loves to read will be fascinated, and teachers will absolutely devour this book."—Daniel Willingham, best-selling author of Why Don't Students Like School?
  • "Language at the Speed of Sight is an incisive tour through the fascinating science of reading. From cuneiform to dyslexia to the future of literacy, Seidenberg is a master guide who--lucky for us--is as gifted a writer as he is a scientist."—Benjamin Bergen, author of What the F

On Sale
Jan 3, 2017
Page Count
384 pages
Basic Books

Mark Seidenberg

About the Author

Mark Seidenberg is the Vilas Research Professor and Donald O. Hebb Professor in the department of psychology at the University of Wisconsin-Madison and a cognitive neuroscientist who has studied language, reading, and dyslexia for over three decades. He lives in Madison, Wisconsin.

Learn more about this author