For an experienced reader, recognising the words on this page feels effortless. Yet, evolutionarily speaking, your brain shouldn’t be able to read. Reading has existed for too short a time to have shaped a dedicated centre in the brain.
Therefore, the brain has to learn to read using systems that evolved for other purposes. But it has long been unclear exactly which systems it recruits to do so and how.
This is important, as differences in these pre-existing systems could shed light on why some of us find reading easier than others. It could also explain how people who are deaf, blind or have dyslexia learn to read in remarkably different ways. In my new paper, published in Cerebral Cortex, I have come up with some answers.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
In my study, I investigated the relationship between individuals’ reading skills and their fine-grained brain structure. I also compared twins with different degrees of genetic overlap to see whether shared genetic factors contribute to the links between brain structure and reading skill.
These patterns offer clues about which parts of the distributed “reading network” might provide the initial conditions for reading, and which might respond more to experience.
Basic, neural hardware
I used high-resolution brain scans from the Human Connectome Project, a large study of healthy young adults. The participants did not read while their brains were being scanned. Instead, they completed a range of language tests separately. The most important of these tests for my study was an adaptive reading test in which participants pronounced written words aloud. I used their scores on this test as a measure of reading ability.
I discovered that several brain areas contribute to the skills needed for reading, and that some show heritable links with it. To be clear, heritability does not mean that we are born a certain way and cannot change. The brain is, in fact, constantly changing. Rather, we begin with a partly inherited lump of clay that we continuously shape and refine throughout our lives.
Let’s begin with auditory abilities. When we learn how to read in school, we typically begin with identifying sounds and putting them together in our minds. In fact, problems with perceiving the sound structure of language are commonly cited as an underlying factor in developmental dyslexia.
To read, children learn to map each letter to a sound and join strings of letters into words: the letter s to the “sss” sound, i to the short “i”, and t to the “t” sound, so that s-i-t is pronounced “sit”. Breaking spoken words into individual speech sounds starts at the syllable level, dividing sitting into sit|ting. This requires the skill to identify slowly developing changes in loudness. Most syllables start quiet with a consonant like s-, then grow louder at the vowel (-i-), with its singing character, before going down in intensity again (-t).
Before we’ve learnt how to read properly, the brain can use auditory cues like these to detect syllables and their internal structure in spoken language and begin its road to literacy.
Particularly suited for this task is a region next to the primary auditory cortex at the top of the left temporal lobe, which sits underneath the temple and upper ear. It is called the medial belt area and is tuned to slow, coarse changes in sound. Strikingly, I found that a well-organized structure of its nerve cells is linked with reading ability.
The relationship had a genetic component, suggesting that the organization of this perceptual system might affect how easy we find it to learn to read. Such differences could contribute to variation in literacy that does not simply separate people with and without dyslexia, but span the whole population.
We also need to understand what we read, however. A skilled reader can usually see words and directly connect them to meaning, in parallel with the activation of their sound. Recognizing the visual aspect of a word is thought to build on our capacity to distinguish faces through their defining features. We identify letters and words by their smaller parts—bows, dots and lines—just like we recognize faces by the eyes, noses and chins.
Face identification seems to depend on a brain area in the right anterior temporal lobe called the dorsal temporal pole. Interestingly, I found that the size of this area is linked with reading ability—and this combination is also passed down. A possible explanation is that it helps us identify written words in a way similar way to how we recognise a face.
Finally, another region in the left anterior temporal lobe (the anterior middle temporal gyrus) is thought to connect word form and meaning. The bigger it is, the better we are at deciphering meaning. And it makes sense that having an advantage in accessing the meaning of words might make it easier to learn to read. Indeed, its size was also genetically related to reading ability.
The role of nurture
Now that we have identified some possible initial conditions for the brain’s reading ability—the lump of clay—we can turn to how the clay might be shaped.
After detecting syllables, the speech-sound distinctions need to be refined. This requires the primary auditory cortex to perform fine-grained sound analysis. Here, a thicker cortex (the outermost layer of the brain) is associated with better reading skills.
But, as I discovered, this combination of reading skill and a thick cortex was not genetically driven. In fact, it might be that experience—including exposure to speech sounds, other forms of auditory analysis (such as picking up different individual sounds from a soundscape) and reading practice—contributes to greater cortical thickness.
The structure of the nerve fibre bundles connecting sound to meaning and visual form also correlates with reading ability. But these relationships do not appear to be heritable either. These bundles might develop when sound, meaning and visual word form are connected during literacy training and general language exposure. So the more we read, the better the brain gets at doing it.
What can we learn from all this? Some brains might have an easier time learning to sound out letters, identify written words or access the meaning of words. In the end, though, our reading and language activities substantially mould the brain’s reading network. In other words, devoted parents, teachers and speech therapists can actually help children develop their inner worlds.
Alternative ways of learning to read
But there is not only one way to learn to read or even only one way to read. I once talked to a physics professor with dyslexia. He was incapable of sounding out words readily and didn’t hear them in his head while reading. He did, however, understand text, describing it as if written words silently built up systems of meaning in his head. He even felt that his reading style was advantageous in his line of work. While listening, I couldn’t keep from wondering what his anterior temporal lobes (involved in identifying familiar forms and accessing meaning) might look like.
My guess is that, because of limited support from the auditory route, his brain has learned to rely more heavily on direct links between written word forms and meaning. His anterior temporal lobes might have helped support this alternative route to understanding what he read.
Of course, that doesn’t mean that dyslexia is a requirement to become a physics professor. For many people, reading difficulties can mean a hard struggle. Even for him, reaching that level of proficiency cannot have been a walk in the park.
People born deaf provide another striking example of the brain finding alternative paths to learn how to read. As with the physics professor, skilled deaf readers can rely heavily on the visual form and meaning of written words. For signers, however, learning to read additionally means building connections between written words and another language—their signed language. Sign languages are independent systems with their own vocabulary and grammar, rather than spoken languages translated into signs.
Braille provides yet another route to reading. It represents letters through patterns of raised dots, allowing words to be read through touch rather than vision. Remarkably, people who become blind early in life can recruit parts of the brain normally devoted to vision when reading Braille.
‘Post-literate’ brains?
Recently, there has been lots of talk about humanity becoming “post-literate”. We increasingly turn to short videos, audio and other media instead of reading longform articles, let alone books. And indeed, an OECD report has found that literacy skills are declining.
We should be worried about this development because literacy ultimately underpins democracy. We need the ability to read at depth, and the continuous attention that comes with it, to properly evaluate ideas and policies.
Regardless of how we read, we all have our clever brains to thank for the ability. We have learned to hack our brains to bring different types of information—auditory, visual and even tactile—together with meaning.
And it doesn’t stop there. We build increasingly complex networks of meaning that help us understand and influence the world we live in.
So, should we surrender to the post-literate era and stop exercising this remarkable brain network? If you ask me—not a chance. Instead, we should keep moulding our grey lumps of clay into ever more beautiful shapes. How do we do that? We read.
