|
|
Reading skills vary widely across the population. Proficient reading brings academic and occupational advantages. Approximately 5-17% of children struggle with learning to read. Functional imaging studies document that reading activates a distributed network of cortical and cerebellar areas. In this talk, I will report on a series of studies that use diffusion magnetic resonance imaging (dMRI) and tractography to characterize white matter pathways in relation to reading ability. dMRI generates a measure called fractional anisotropy (FA), which assesses tissue properties that affect water diffusion in the brain. Higher FA means greater restriction in diffusion, which could arise from increased myelin, increased axonal packing or larger axonal diameter.
In study 1, we compared FA along major white matter pathways in 6-year old children who could decode written pseudowords (Readers) to age-matched peers who could not (Pre-Readers). We found that Readers had significantly increased FA along the left superior longitudinal and the right uncinate fasciculus compared to Pre-Readers. Interestingly, variation in the right uncinate was associated with language skills and with reading. In study 2, we asked whether variations in FA at age 6 would predict reading at age 8. We learned that mean tract FA of the left and right superior longitudinal fasciculus predicted individual differences in reading, beyond the contribution of other known predictors of reading, including demographic variables, language skills, and phonological awareness. In study 3, we analyzed FA of the three cerebellar peduncles that connect the cerebellum to the cortex, thalamus, and brainstem. FA of all cerebellar peduncles correlated with scores for both decoding and comprehension. In addition, FA of the left inferior cerebellar peduncle at age 6 was a unique predictor of reading ability at age 8. In study 4, we analyzed quantitative MRI scans to obtain a proxy measure for myelin content of white matter pathways. We unexpectedly found that reading did not correlate with myelin content, but rather did correlate with measures of tract cohesion derived from dMRI. In study 5, we replicated these methods with children born preterm. Patterns of association and prediction were not the same as in children born full term.
In summary, white matter matters in reading. FA of white matter pathways connecting long range cortical areas and linking cortex to cerebellar areas are associated with individual differences in reading skills. However, the neurobiology of reading may vary in different clinical populations. Future studies should consider whether reading experience or remediation can change both white matter microstructure and reading outcomes in children with reading disorders, with or without other clinical conditions.
|
|