Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Nature and nurture in the cortical microstructure of reading
Poster Session F, Friday, October 2, 2:45 - 4:45 pm, Wangari Maathai
Mikael Roll1; 1Lund University
Introduction: Reading acquisition builds on interacting phonological and semantic skills. However, the microstructural cortical correlates of these abilities remain poorly understood. This study investigated the relationship between reading skill and cortical and white matter microstructure. Functionally defined cortical subregions and white matter tracts involved in phonological and semantic processing[1] were analyzed in a large normative sample: 1,069 Human Connectome Project Young Adult (HCP-YA) participants.[2] Further, heritable and non-heritable cortical correlates of reading across the auditory and semantic systems were dissociated in a genetically confirmed twin subsample. Methods: Cortical thickness, surface area, and T1w/T2w ratio were assessed for auditory and semantic subregions using a multimodal cortical parcellation[3] of HCP-YA. T1-weighted images were used for cortical thickness and surface area. The parcellation and T1w/T2w ratio were also based on T2-weighted images. T1w/T2w ratio was used as a proxy for cortical myelin content. Diffusion metrics were extracted using the XTRACT atlas on preprocessed data. Reading ability was assessed using the NIH Toolbox Oral Reading Recognition Test, which evaluates letter and word pronunciation through an adaptive procedure. Heritability and genetic correlation were estimated using bivariate Cholesky decomposition ACE (additive genetic, common environment, unique environment) models[4] (monozygotic = 246, dizygotic = 152). Results: Microstructural traits in auditory and anterior temporal subregions predicted oral reading performance. In the left auditory cortex, T1w/T2w ratio in a paralemniscal belt area showed heritable associations with reading, whereas thickness in the lemniscal core region reflected non-heritable contributions. Bilateral ATL surface areas were genetically associated with reading and independently predicted semantic performance. Left ventral fiber bundles and a left ATL region, connected to left dorsal fibers, predicted reading scores independent of genetic influences. The associations were continuous across the full range of reading ability, supporting dimensional accounts of reading-related cortical variation. Discussion: Previous findings of cortical thinning[5] and hypermyelination[6] in the auditory cortex as defining traits of persons with dyslexia can now be generalized to a continuous relationship with declining reading skills in a larger sample of typical readers. The genetic relationship between hypermyelination and reading may originate in altered low-frequency auditory sampling, a precursor to phonological awareness,[7] possibly due to variation in neuronal migration during gestation.[8] Experience-driven auditory core thickness could arise from compensatory increases in the use of fine-grained acoustic cues to detect speech sound boundaries. The genetic associations between anterior temporal surface areas and oral reading are likely to stem from increased semantic capacity facilitating word learning. Among white matter tracts, ventral fibers showed the strongest associations with the word-level reading task, consistent with a contribution of semantic processing. References: [1] Roll M. Neuroimage 303, 120930 (2024). [2] Van Essen DC et al. Neuroimage 80, 62-79 (2013). [3] Glasser MF et al. Nature 536, 171-181 (2016). [4] Neale MC, Cardon LR. (Kluwer Academic Publishers, 1992). [5] Clark KA et al. Brain 137, 3136-3141 (2014). [6] Skeide MA et al. Neurology 90, e492-e497 (2018). [7] Goswami U. Trends Cogn. Sci. 15, 3-10 (2011). [8] Giraud A-L, Ramus F. Curr. Opin. Neurobiol. 23, 37-42 (2013).
Topic Areas: Reading, Genetics