The Neuroscience of Reading: A Skill Your Brain Was Never Built For

A man who had never learned to read agreed to lie inside a brain scanner while researchers showed him words. Nothing happened. The words went in as shapes and stayed shapes. Then he started literacy classes, and they scanned him again, and again over the following months. Slowly, in a specific patch of tissue on the underside of his brain, something switched on that had not been there before.

Alongside it, something else dimmed. In the same neighborhood, the response to human faces began to fade.

A skill with no organ

Here is why that happens. Vision is millions of years old. Speech is old enough that a child soaks it up from the people around them without a single lesson. Writing is about five thousand years old, which in evolutionary terms is last Tuesday. There was never time to build a reading organ, and there is no gene for it.

Every person who reads has taken a piece of brain that was doing another job and put it to work on letters. Stanislas Dehaene and Laurent Cohen named the process neuronal recycling, and the region it happens in is called the visual word form area, tucked in a fold on the left underside of the brain.

The strange part is that it appears in the same place in almost everybody. Portuguese, Brazilian, Chinese, Hebrew readers, the same fold. Reading does not land wherever there is room. It takes the piece of cortex already good at fine detail in the center of vision, at line junctions and small shapes, because that hardware was closest to the work.

What sixty-three adults showed

In 2010 Dehaene’s team published a study in Science that remains the clearest picture of this. They scanned sixty-three adults in Portugal and Brazil, sorted into three groups: ten who had never learned to read, twenty-two who learned as adults, and thirty-one who learned as children.

The line that matters most for anyone who thinks they left it too late: most of these changes appeared in the adults who learned to read in adulthood. A brain in its forties or fifties reorganized itself much the way a child’s does. Slower, harder, and it still happened.

The thing you gave up without noticing

Now for the part almost nobody knows.

Imagine a lion. Now imagine it facing the other way. Same lion, obviously. Every visual system in nature works this way, treating a shape and its mirror image as one thing, because for a few hundred million years that was the correct answer. A predator coming from the left is the predator coming from the right.

Adults who have never learned to read still live in that world. Show them a shape and its mirror twin and ask which is which, and they find it genuinely hard. Their ability to see through the flip is actually better than yours.

Then somebody hands them an alphabet, where b and d are one shape flipped, and p and q are the same trick again. To read English you have to break a rule your eyes have followed since before your species existed.

Why children reverse their letters

A child writes her name with the E facing backward and somebody at the school suggests she be assessed.

Nothing is wrong with her. She is doing what eyes have always done. Her visual system is still handing her the ancient answer, that a shape and its mirror are the same shape, and she has not yet finished overriding it. The override is the unnatural part, and it takes years.

The same child, a different alphabet

Now move that child to Beijing.

Chinese characters do not spell sounds. Each one packs meaning and a syllable into a dense square of strokes that has to be held whole in the mind. Reading Chinese leans hard on a different region, the left middle frontal gyrus, which handles verbal and spatial working memory.

In 2004, Wai Ting Siok and colleagues published a paper in Nature with a title that says it outright: the biological abnormality of impaired reading is constrained by culture. Children who struggle to read Chinese show disruption in that middle frontal region. Children who struggle to read English show it somewhere else entirely, in temporoparietal and occipitotemporal areas. A follow-up in 2008 found less gray matter in the Chinese-specific region in struggling Chinese readers.

Later research has muddied this a little, in a useful direction. Some studies find deficits shared across Chinese and English readers, or across Chinese and French, sitting alongside the script-specific ones. The current picture is convergent and divergent at the same time.

What holds up is the human conclusion. A reading difficulty is not simply a fact about a child. It is a fact about a child meeting one particular writing system, and the same child in a different script would be handed a different set of problems, and possibly a different diagnosis.

What this means at your kitchen table

Nothing about reading is automatic, ever, for anyone. Speech arrives on its own in any household where people talk. Reading has never arrived on its own in any culture on earth. Somebody always has to be shown.

• Backward letters are on schedule. The mirror habit has to be actively suppressed, and suppression is slow.
• Slow does not mean unintelligent. The task is genuinely against the grain, and the timetable varies enormously between children.
• Volume beats difficulty. That patch of cortex tunes itself through repetition. An easy book finished does more for it than a hard book abandoned.
• Let them reread. The forty-third time through the same picture book is the circuit getting faster.
• Do not compare across languages. Different scripts make different demands and produce different mistakes.

Go back to the man in the scanner. He was an adult, unschooled, and by every folk theory about brains he had missed his window. His brain built the region anyway, and the researchers watched it happen month by month. What the scans could not show is what it felt like on the morning he first read a sign in the street and understood it before he could stop himself.