Neural precursors of math learning disability among children in kindergarten – PREMATH

Mathematical Learning Disability (also termed dyscalculia) is a difficulty in learning math that is not explained by abnormal cognitive functioning, sensory deficits, or inadequate schooling. Math Learning Disability (MLD) affects 5% to 10% of children worldwide and -- given the importance of math skills in our technology- and information-driven society -- has wide-ranging consequences for both affected individuals and nations. Only a few neuroimaging studies have investigated the functional and anatomical brain correlates of MLD. These have generated somewhat inconsistent results, with brain differences between children with and without MLD encompassing the parietal and prefrontal cortices, as well as the medial temporal lobe and occipital gyrus. All of these studies, however, have been conducted on children who were already experiencing math difficulties. Therefore, it is unknown whether these abnormalities result from experiencing math failures or predate these difficulties. This project has two main goals. The first is to determine what brain abnormalities are present in 5-year-old children (children in kindergarten) who are at risk of developing MLD but have not been exposed to most of their arithmetic instruction in elementary school (and are therefore not diagnosed yet). The second goal is to test whether brain measures of children at risk for MLD in kindergarten can be used to predict which children will actually progress into MLD in 2nd grade. Children will be recruited using a two-step process. First, over two years, 2,000 children in the last months of pre-K from 33 public schools in the Lyon area in France will be screened using a two-minute paper-and-pencil test of symbolic and nonsymbolic numerical magnitude processing. This test is related to individual differences in children's calculation skills. Second, children who will be identified at risk for MLD based on the paper-and-pencil test will be invited to the lab during their kindergarten year for a more extensive assessment of early math abilities with standardized tests. Children with no risk of MLD will also be recruited using the same procedure. All children will then undergo both structural and functional magnetic resonance imaging (MRI) scanning. Voxel-based morphometry, activity and connectivity analyses will be used to identify brain differences between children with and without risk of MLD in kindergarten. All children will then be longitudinally followed to identify which at-risk children actually progressed to MLD in 2nd grade. We will use multivariate statistics to test whether fMRI data collected two years earlier could have been used to predict which at-risk children would progress to MLD. The project has implications for both improving our understanding of the neural impairments leading to MLD, as well as the detection of children who are at risk of MLD as early as in kindergarten.