No evidence for systematic white matter correlates of dyslexia: An Activation Likelihood Estimation meta-analysis

•Previous literature indicates important discrepancies in structural differences associated with dyslexia.•We explore the relationship between dyslexia, reading and fractional anisotropy, focusing on voxel-based analyses.•Our meta-analysis shows no reliable clusters underlying differences between dyslexics and typical individuals.•A second analysis showed that reading ability was not associated with reliable differences in white matter integrity.•These findings do not support the idea of systematic differences in FA between dyslexic and typical readers. Dyslexia is a prevalent neurodevelopmental disorder, characterized by reading and spelling difficulties. Beyond the behavioral and functional correlates of this condition, a growing number of studies have explored structural differences between individuals with dyslexia and typically developing individuals. To date, findings remain disparate – some studies suggest differences in fractional anisotropy (FA), an indirect measure of white matter integrity, whereas others do not identify significant disparities. Here, we synthesized the existing literature on this topic by conducting a meta-analysis of Diffusion Tensor Imaging (DTI) studies investigating white matter correlates of dyslexia via voxel-based analyses (VBA) of FA. Our results showed no reliable clusters underlying differences between dyslexics and typical individuals, after correcting for multiple comparisons (false discovery rate correction). Because group comparisons might be too coarse to yield subtle differences, we further explored differences in FA as a function of reading ability, measured on a continuous scale. Consistent with our initial findings, reading ability was not associated with reliable differences in white matter integrity. These findings nuance the current view of profound, structural differences underlying reading ability and its associated disorders, and suggest that their neural correlates might be more subtle than previously thought.