Poster A61, Thursday, August 16, 10:15 am – 12:00 pm, Room 2000AB

An Investigation of the Relationship between Reading and Speech Production in Children and Adolescents

Cassidy Fleming1, Angela Cullum1, Jacqueline Cummine1,2;1Faculty of Rehabilitation Medicine, University of Alberta, 2Neuroscience and Mental Health Institute, University of Alberta

Purpose. Approximately 10-15% of children have reading difficulties, including developmental dyslexia and specific comprehension deficit. Relationships between basic reading processes and brain structures in adults have been widely reported in neuroimaging literature. However, few studies have examined these relationships in children, and many face limitations of small sample size. The primary objective of this study is to examine the relationship between structural characteristics of the brain and reading performance across the developmental trajectory by using a combination of diffusion tensor imaging (DTI; i.e., fractional anisotropy), and volumetric magnetic resonance imaging (MRI; i.e., volume and thickness) measures. Method. Child and adolescent participants aged 3 to 17 (n=444) were obtained from the Pediatric Imaging, Neurocognition and Genetics (PING) Study database. Participants were included if they were 6 to 17 years of age, right handed, native English speakers, had completed the NIH Toolbox Oral Reading Recognition Test (TORRT), did not have vision or hearing problems and did not have a diagnosis of attention-deficit hyperactivity disorder. Volume, thickness and fractional anisotropy of brain regions and white matter tracts implicated in reading and speech production were extracted from the PING Study database, along with TORRT reading scores. Participants were divided into four age groups (years;months): 3;0 to 6;11, 7;0 to 11;5, 11;5 to 13;11 and 14;0 to 17;11. Blockwise multiple linear regression analysis was used to examine the relationship between age-corrected TORRT reading scores and DTI and MRI measures by age group, while controlling for potential confounds including prenatal alcohol exposure, parental education and occupation, household income, and racial minority status. Results. The neural correlates of reading performance vary significantly across development and reflect the demands of reading across the developmental trajectory. The regression model for each age group accounts for between 9.9% and 33.9% of the variance in reading scores. Regions of the brain associated with articulatory representations have a significant influence on reading at every stage of development, including the pars triangularis, cerebellum and putamen. Significant changes continue to occur throughout late childhood and adolescence. Significance. The PING study database provided a unique opportunity to examine neural correlates of reading performance in a large cohort of children and adolescents. The findings of this study expand our understanding of brain-behavior relationships in reading, and the influence of articulation on reading, providing evidence to support the print-to-speech model. Finally, these findings have the potential to inform optimal treatment of individuals with reading disabilities.

Topic Area: Language Development