Brain Structure Correlates of Component Reading Processes: Implications for Reading Disability

Abstract

Brain structures implicated in developmental dyslexia (reading disability - RD) vary greatly across structural magnetic resonance imaging (MRI) studies due to methodological differences regarding the definition of RD and the exact measurements of a specific brain structure. The current study attempts to resolve some of those methodological concerns by examining brain volume as it relates to components of proposed RD subtypes. We performed individual regression analyses on total cerebral volume, neocortical volume, subcortical volume, 9 neo-cortical structures and 2 sub-cortical structures. These analyses used three dimensions of reading, phonemic ability (PA), orthographic ability, and rapid naming (RN) ability, while accounting for total cerebral volume, age, and performance IQ (PIQ). Primary analyses included membership to a group (poor reader vs. good reader) in the analysis. The result was a significant interaction between PA and reading ability as it predicts total cerebral volume. Analyses revealed that poor readers lacked a relationship between PA and brain size, but that good readers had a significant positive relationship. This pattern of interaction was not present for the other two reading component factors. These findings bring into question the general belief that individuals with RD are at the low end of a reading ability distribution and do not have a unique disorder. Additional analyses revealed only a few significant relationships between brain size and task performance, most notably a positive correlation between orthographic ability and the angular gyrus (AG), as well as a negative correlation between RN ability and the parietal operculum (PO).