The sharing of neuroimaging data is crucial to allow for reproducible and reliable cognitive neuroscience. Neuroimaging data of academic skills is important because it informs cognitive models of reading and arithmetic, but also because these skills are crucial for success in our society. This project will archive data from three large-scale neuroimaging studies of math and reading development, with up to over 150 participants in the cross-sectional component and 50 participants in the longitudinal component. The longitudinal component allows for the examination of growth trajectories and whether these can be uniquely predicted from early neural data. The studies investigate 8- to 15-year-old children with functional magnetic resonance imaging (fMRI), structural MRI and diffusion tensor imaging (DTI). These different imaging modalities allow for sophisticated analyses of the interrelationship between brain structure and function. Moreover, all studies involve multiple tasks, with one using a multisensory design to examine the lexicality effect by comparing words to pseudowords, another using verbal and spatial localizers to identify different mechanisms in arithmetic processing, and a third employing multiple tasks to investigate orthographic, phonological and semantic processing. Moreover, there are manipulations within each task to allow for parametric statistical models. Item-level task performance and stimulus characteristics will also be provided to allow for psycholinguistically motivated analyses of the data. In addition, all projects have a partially overlapping, extensive psycho-educational standardized testing battery as well as a medical and developmental history questionnaire. We will use the state-of-the-art OpenfMRI portal to upload our data. OpenfMRI allows for unrestricted access of raw data in the standardized Brain Imaging Data Structure (BIDS) format, and provides analytical tools to work with the data. Harmonization with other shared data sets will allow for new questions to be asked, like whether there are domain-specific or domain-general mechanisms in academic skills. There are two aims of the project: #1 Digital curation, and #2 Documentation and dissemination.
The first aim will involve de-identifying the neuroimages through defacing, checking for data quality through movement and artifact detection, organizing data according to the BIDS format, and finally, the submission to a data curator for the calculation of more specific quality metrics.
The second aim will involve documentation of the materials and procedures used in each experiment including experimental task, standardized testing, questionnaires and MRI acquisition parameters.
The second aim will also involve dissemination through the formulation of three papers to be submitted to Data in Brief.
This project will archive neuroimaging data from three large-scale projects on children. Deeper understanding of the development of and individual differences in the neural mechanisms of reading and math will have implications for the identification of and intervention in learning disabilities (i.e. dyslexia and dyscalculia).