The science of reading has roots stretching back more than a century, but the term itself gained widespread public attention only in the late 2010s. The foundational research began in 1908, when psychologist Edmund Burke Huey published a pioneering book examining how the eyes and brain work together during reading. What followed was a slow, uneven accumulation of cognitive research, theoretical models, and federal reports that collectively built the body of knowledge now called “the science of reading.”
The 1908 Book That Started It All
Edmund Burke Huey’s “The Psychology and Pedagogy of Reading,” published in 1908, is widely considered the first serious scientific examination of how people read. Huey treated reading as an information-processing activity, one where an arbitrary set of symbols transfers meaning from one mind to another. He described how the eyes move across text, examined what the reader actually perceives on a printed page, and then speculated about the higher-level cognitive operations that translate those visual impressions into meaning. His work framed reading not as a simple skill but as a complex chain of mental events, from eye movement to perception to comprehension.
For decades after Huey, reading research continued in university psychology departments, but it remained largely academic. Classroom instruction was shaped more by tradition and philosophy than by cognitive science. That gap between what researchers knew and what teachers did would persist for most of the 20th century.
Key Models From the 1980s and 1990s
The field accelerated in the late 1980s and 1990s as researchers proposed specific frameworks for how reading works in the brain. In 1989, Mark Seidenberg and Jay McClelland introduced a connectionist model describing four processors the brain uses during reading: an orthographic processor for recognizing letters, a phonological processor for recognizing sounds, a meaning processor, and a context processor. This model showed reading as an interaction between visual, auditory, and language systems rather than a single skill.
In 1990, Wesley Hoover and Philip Gough formalized what became known as the Simple View of Reading. Their framework proposed that skilled reading requires just two components: decoding (the ability to translate print into speech sounds) and linguistic comprehension (the ability to understand language). Both are necessary. A child who can decode words but doesn’t understand their meaning isn’t truly reading, and neither is a child who understands spoken language but can’t decode print. This simple equation gave teachers and researchers a clear way to diagnose where a struggling reader was breaking down.
In 2001, Hollis Scarborough added further nuance with her Reading Rope model, which visualized skilled reading as multiple strands woven together. The word recognition side included phonological awareness (recognizing the sounds within words), decoding and spelling, and automatic sight recognition of familiar words. The language comprehension side included background knowledge, vocabulary, understanding of sentence and text structure, verbal reasoning, and literacy knowledge, meaning familiarity with how different types of texts are organized. As readers develop, these strands become increasingly automatic and intertwined, producing fluent comprehension.
The 2000 National Reading Panel Report
The single most influential policy moment in the science of reading’s history came in 2000, when the U.S. Congress commissioned the National Reading Panel to review decades of research and determine what actually works in reading instruction. The panel identified five essential components supported by evidence: phonemic awareness (recognizing and manipulating individual sounds in words), phonics (connecting sounds to letters), fluency, vocabulary, and comprehension strategies. These became known informally as the “five pillars” of reading instruction.
The report was a landmark because it translated a sprawling body of academic research into concrete instructional recommendations. It also made explicit what many researchers had argued for years: that systematic phonics instruction produces better outcomes than approaches that treat reading as a naturally acquired skill. Still, the report’s findings were unevenly adopted. Many teacher preparation programs and school districts continued using methods that contradicted the evidence.
What Neuroscience Revealed About the Reading Brain
Brain imaging research added a biological dimension to the science of reading. Neuroscientist Stanislas Dehaene described a process called “neuronal recycling,” in which a region of the brain’s visual cortex, originally evolved for recognizing faces and objects, gets repurposed for letter recognition when a person learns to read. This region reliably appears in the same location across literate brains, likely because it has preexisting connections to areas involved in spoken language processing.
This finding carried a profound implication: the human brain is not wired to read. Unlike spoken language, which children acquire through exposure, reading is an invention that must be explicitly taught. The brain has to physically reorganize itself, recruiting visual hardware and linking it to language circuits. In literate brains, written words end up encoded by neurons tuned to individual letters, their positions within words, and common letter pairs. This biological reality reinforced what decades of behavioral research had shown: children need direct, systematic instruction to learn how print maps onto speech.
The 2018 Public Reckoning
For most of its history, the science of reading was a conversation among researchers, cognitive scientists, and a relatively small group of literacy advocates. That changed in September 2018 when journalist Emily Hanford published “Hard Words,” an investigative report for APM Reports. Hanford’s central argument was blunt: scientific research had shown how children learn to read and how they should be taught, but many educators didn’t know the science and, in some cases, actively resisted it. Millions of kids were being set up to fail.
The report highlighted a fundamental disconnect. The prevailing approach in many American schools assumed that learning to read is a natural process, much like learning to talk. But that assumption contradicted decades of research showing the opposite. Children must be explicitly taught how to connect sounds with letters. Hanford’s reporting reached an enormous audience, and the phrase “science of reading” entered mainstream conversation in a way it never had before.
The Legislative Wave That Followed
Hanford’s reporting, combined with growing parent frustration and persistently low reading scores, triggered a wave of state legislation. From 2012 through 2018, only one or two states passed science of reading legislation in any given year. In 2019, eight states passed bills. After a pause in 2020, the pace exploded: 13 states in 2021, eight in 2022, and 17 in 2023. By 2024, at least 35 literacy bills had been enacted across 25 states in a single year. In total, 38 states and Washington, D.C., have now passed legislation related to evidence-based reading instruction.
These laws typically require schools to use phonics-based curricula, mandate new training for teachers, and sometimes ban the instructional methods (like three-cueing, which encourages children to guess words from context or pictures) that the science has shown to be ineffective. The legislative movement represents the first time the century-old body of reading research has been translated into widespread classroom practice through policy rather than persuasion alone.
A Century of Research, a Decade of Action
The science of reading didn’t begin at a single moment. Its roots are in Huey’s 1908 observations about eye movements and cognition. Its theoretical backbone was built through models developed in the 1980s and 1990s. Its policy foundation was laid by the National Reading Panel in 2000. Its biological basis was confirmed by brain imaging studies in the 2000s and 2010s. And its public breakthrough came with investigative journalism in 2018. The research accumulated over more than a hundred years, but the political and educational will to act on it is largely a phenomenon of the last five.