A recent large-scale study published in Science Advances has revealed a connection between genetic variations associated with dyslexia and structural differences in the brain. These differences were found in areas involved in motor coordination, vision, and language. This provides new insights into the neurological underpinnings of this common learning difficulty.
Dyslexia is a common learning difficulty that primarily affects the skills involved in accurate and fluent word reading and spelling. It’s characterized by challenges with phonological awareness (the ability to recognize and manipulate the sounds in spoken language), verbal memory, and verbal processing speed. People with dyslexia may struggle to decode words, recognize familiar words automatically, and spell words correctly. Importantly, dyslexia is not related to a person’s overall intelligence. It’s considered a neurodevelopmental condition, meaning it arises from differences in how the brain develops and processes information, particularly related to language.
Dyslexia is known to have a strong genetic component, often running in families. Twin studies have estimated that 40% to 70% of the variation in dyslexia risk can be attributed to genetic factors. While previous research has identified some brain regions that may function differently in individuals with dyslexia, these findings have often been inconsistent, possibly due to small sample sizes and differences in study methods. Additionally, the precise relationship between specific genes associated with dyslexia and their impact on brain structure has remained unclear.
The motivation behind the new study was to address these gaps in our understanding by leveraging the power of large-scale data. The researchers recognized that investigating the connection between genetic predisposition to dyslexia and brain structure in a very large sample could provide more robust and reliable insights than smaller, more traditional studies. They aimed to identify specific brain regions and white matter tracts that are associated with genetic risk for dyslexia, and to explore whether different genetic variants might influence distinct neural pathways.
“Thirty-five genetic variants that influence the chance of having dyslexia were already known from a very large study by the company 23andMe in the USA, carried out in over one million people. However, that study did not include brain MRI data. The new aspect of our study was to investigate the genetic variants in relation to brain structure in MRI data from thousands of people,” explained Clyde Francks (@clydefrancks), a professor at the Max Planck Institute for Psycholinguistics in Nijmegen and senior author of the study.
The researchers used two large datasets: the genetic data 23andMe and brain imaging data from over 30,000 adults in the UK Biobank. The 23andMe dataset helped identify genetic variants associated with dyslexia by comparing individuals who reported a dyslexia diagnosis to those who did not. These genetic variants were then used to calculate “polygenic scores” for individuals in the UK Biobank, reflecting their genetic predisposition to dyslexia.
Although the UK Biobank participants were not specifically diagnosed with dyslexia, their polygenic scores varied, allowing researchers to analyze how this genetic predisposition related to differences in brain structure. Brain scans from the UK Biobank were examined to assess differences in regional brain volume and white matter microstructure—features that provide insight into how different areas of the brain are connected and function.
To refine their analysis, the researchers used a statistical technique known as independent component analysis. This allowed them to separate out different patterns of brain structure associated with specific genetic variations, helping them identify distinct “impact modes”—sets of genetic variants linked to unique structural features in the brain.
“The genetic contribution to dyslexia involves many thousands of genetic variants all across the genome, each with a small effect on the chance of having dyslexia, but in combination they add up to a measurable contribution,” Francks said.
The researchers found that individuals with a higher genetic predisposition to dyslexia tended to have lower overall brain volume, with the effect being more pronounced in gray matter than in white matter. More specifically, genetic risk for dyslexia was linked to reduced volume in several brain regions, including the medial frontal cortex, midbrain, thalamus, and bilateral amygdalae. These areas are associated with higher-level cognitive functions, attention, and language processing.
“Our study implicated various brain regions and networks linked to the genetic chance of having dyslexia, which were involved prominently in motor coordination, language and vision, although not limited to those functions,” Francks explained. “This does not mean that every person with dyslexia has changes in all of these brain systems. Rather, it is likely that some people with dyslexia have a particular genetic contribution affecting for example their motor functions, others their language functions, and others their vision functions, while some people may have different combinations of these. Part of our study was aimed at breaking down the overall genetic chance of having dyslexia into distinct components that associate with different brain networks.”
Certain structural differences were particularly notable. For example, individuals with a higher genetic risk for dyslexia had reduced volume in the left temporoparietal junction and the left anterior insula—areas known to play critical roles in language and phonological processing. This supports the idea that dyslexia is linked to difficulties in recognizing and processing the sounds of language.
The researchers also found that polygenic scores for dyslexia were linked to changes in the structure of white matter pathways. Specifically, individuals with a higher genetic risk for dyslexia showed increased white matter density in the forceps major—a tract connecting the two occipital lobes, which are crucial for visual processing.
Meanwhile, reduced white matter density was observed in pathways connecting the cerebellum to the cortex, including the superior longitudinal fasciculus and the anterior limb of the internal capsule. These findings align with previous research suggesting that dyslexia is associated with motor coordination difficulties, as well as differences in how visual and linguistic information is integrated.
Interestingly, the study also examined whether the brain regions associated with dyslexia-related genes overlapped with those linked to other cognitive traits. The results showed substantial overlap with brain regions associated with intelligence, educational attainment, and attention-deficit hyperactivity disorder (ADHD). However, one region stood out as being uniquely linked to dyslexia: the primary motor cortex. This finding suggests that motor function might play a particularly important role in dyslexia, distinguishing it from other cognitive traits.
“There are various traits that are partly associated with dyslexia in the population, including educational attainment and attention deficit/hyperactivity disorder,” Francks said. “There is evidence that certain genetic variants impact dyslexia as well as these other traits. This is why we looked at how genetic effects on these other traits are related to brain structure.”
“In this way we could assess which aspects of brain structure are linked relatively specifically to the genetic chance of having dyslexia, as opposed to more generally to genetic effects on other related traits. For example, lower motor cortex volume was relatively specific to the genetic chance of having dyslexia, whereas lower nerve fiber density in the internal capsule was found more generally in relation to genetic effects on dyslexia, ADHD, educational attainment and fluid intelligence.”
The study provides important insights into the neural correlates of genetic predisposition to dyslexia. But there are some limitations. The UK Biobank sample, while large, may not be fully representative of the general population due to volunteer bias. The dyslexia polygenic scores were based on self-reported diagnoses in the 23andMe study, without detailed information on the type or severity of dyslexia. Additionally, the study was cross-sectional, meaning it examined data at a single point in time. This makes it difficult to determine whether the observed brain differences are a cause or consequence of dyslexia.
Future research should aim to replicate these findings in other large samples, ideally including longitudinal data from children to track brain changes during reading development. Investigating potential sex differences in the genetic and neural underpinnings of dyslexia would also be valuable.
In the future, polygenic scores, combined with other factors like early cognitive assessments, might help tailor education to individual needs. However, larger genetic studies of dyslexia are needed to improve accuracy.
“Maybe in the future polygenic scores will be accurate enough, when used in combination with other types of data (such as pre-school cognitive and behavioural assessments), to have a useful impact on adjusting education to a child’s particular needs,” Francks said. “For the time being this is not possible. Even larger genetic studies of dyslexia would first need to be carried out, so that effects of each genetic variant can be measured more precisely.”
“If there comes a point in the future when all infants are routinely genotyped as part of standard healthcare assessment, then applications of polygenic scores might become feasible also in the educational domain. People would of course also need to decide whether this is desirable. In terms of feasibility, there would need to be enough of a predictive gain at the individual level, although cognitive and behavioural assessments will remain the most important.”
The study, “Distinct impact modes of polygenic disposition to dyslexia in the adult brain,” was authored by Sourena Soheili-Nezhad, Dick Schijven, Rogier B. Mars, Simon E. Fisher, and Clyde Francks.
