De ontwikkeling van het neuro-anatomisch leesnetwerk bij kinderen met een verhoogd risico op dyslexie

Although our brain is not predestined for reading, as alphabetic scripts exist for around 3000 years, the vast majority of the population learns to read fluently, presumably by reorganizing the already established visual and language brain network. However, 3 to 7% of the children do not learn to read fluently, and they develop dyslexia. According to the multiple deficit model of Pennington (2006) the aetiology of developmental dyslexia is multifactorial. Hence, developmental dyslexia is thought to result from impairments at different levels, i.e. the genetic, neurobiological, environmental and cognitive level, and from the interactions between these levels. Previous neuroimaging studies gained insights into the neural correlates of mature reading, showing that reading generally involves a widespread network of distinct regions in the left hemisphere of the brain. To date, it remains unclear how the brain reorganizes throughout the first stages of reading acquisition, and whether deviances in children with (a familial risk for) dyslexia are present prior to reading onset or emerge during reading acquisition. By three longitudinal studies (Chapters 2, 3, 5) and one methodological oriented study (Chapter 4), the main anatomical region considered to host phonemic representations, i.e. the planum temporale, and the most relevant white matter pathways are studied in relation to reading and to different levels of the multiple deficit model. In the first study (Chapter 2), we investigated planum temporale surface area in pre-readers and in adolescents. The results revealed atypical lateralization of the planum temporale in relation to a family history of dyslexia, rather than to the (later) reading level. In the second study (Chapter 3), we applied diffusion tensor imaging (DTI) to investigate previously reported white matter anomalies in adults with dyslexia, in a longitudinal sample of pre-readers and early readers. We demonstrated that white matter deviances in the dorsal arcuate fasciculus are dynamic and precede reading acquisition, and moreover, ventral white matter deviances were detected in relation to a family history of dyslexia. The latter deviances were stable over time. Finally, we aimed at investigating reading induced structural brain changes, especially with regard to phonological processing and orthographic knowledge. Therefore, the higher order non-tensor Spherical Deconvolution (SD) model was applied. The results revealed that the relation between dorsal white matter and phonological processing precedes reading acquisition, and this relation is specific to the arcuate fasciculus (Chapter 4). The specialization of ventral white matter for orthographic knowledge, especially in the left inferior fronto-occipital fasciculus, emerges throughout the early stages of reading acquisition (Chapter 5). The results obtained within this doctoral dissertation are further discussed in a general discussion (Chapter 6). In sum, we demonstrate that neuroanatomical deviances can precede reading acquisition and that neural reorganization parallels the initial stages of reading development. While some deviances seem to relate to the emergence of developmental dyslexia, other deviances are merely related to a familial risk for dyslexia. These results emphasize the complex aetiology of developmental dyslexia.status: publishe