Assessing the Relationship Between Relational Memory and Segregation of Intrinsic Functional Brain Networks Vulnerable to Alzheimer’s Disease During Periadolescent Development: Preliminary Findings from the PRANK Study

AbstractBackgroundAlzheimer’s disease (AD) is an insidious disease process that is frequently diagnosed following complaints of memory loss. The AD‐related decline in memory abilities has been attributed to functional and pathological changes in the hippocampus and distributed functional networks supporting hippocampal‐dependent memory. Although AD manifests in late life, lifelong risk factors for AD may be influenced by childhood brain development. The default mode network (DMN) is one functional network associated with the hippocampus and has been well‐characterized in adulthood. However, there is a gap in the literature regarding the connection between DMN network properties and relational memory abilities in childhood. Here we report preliminary findings from our ongoing study (Polygenic Risk of Alzheimer’s disease in Nebraska Kids, PRANK). The PRANK study collects functional MRI, cognitive measures, and AD polygenic risk scores from periadolescent children in Nebraska (age 8‐13 years). Here, we report the current findings measuring the relationship between brain network measures and cognitive outcomes from the PRANK study.MethodTo assess the relationship between intrinsic functional networks of the brain and hippocampal‐dependent memory, we applied brain and cognitive measures from a preliminary sample of periadolescent children (N=90) from the PRANK study. Cognitive assessments measured hippocampal‐dependent memory and other cognitive domains. Brain measures included both structural and resting‐state functional MRI data. Two properties of brain networks were measured, modularity and participation coefficient. These measures were derived from resting‐state MRI data with help of the Human Connectome Project’s Connectome Workbench tool. For comparison, brain and cognitive measures were also collected from a separate dataset of healthy young adults.ResultOur analysis observed differences between the PRANK and young adult datasets. The observed between‐group differences included graph theoretic measures of network architecture, particularly modularity. The relationship between relational memory performance and the network properties PRANK participants was also assessed.ConclusionThe ongoing PRANK study seeks to determine the association of brain measures and cognitive performance. Here, our preliminary analysis measured the relationship between brain and cognitive measures in periadolescent children with a focus on brain network properties and relational memory.