Quantifying individual differences in patterns of functional brain network organization in youths

During childhood, the brain undergoes rapid periods of development that are vital for the cognitive processes supporting goal directed behavior. Some of these processes are termed ‘executive functions’, and the development of these skills not only impacts academic achievement, but also lifelong success. Successful navigation of executive function relies on the specialized integration of brain regions, thus requiring a network- level view of brain organization to better understand associations with behavioral outcomes. This dissertation examines multiple methods of quantifying patterns of brain activity at rest and tests the association of these measures with individual differences in youth executive function task performance. Study 1 identified patterns of brain activity unique to an individual (termed resting state neural fingerprints) in both youths and adults, using support vector machine classifiers. We found that the classifiers successfully identified an individual’s scan from their own previous scan and one twin’s scan from their co-twin’s scan. Our results suggest that resting state functional fingerprints are stable over time and vary by genetic relatedness. Study 2 quantified patterns of resting state functional network organization in youths, using graph metrics chosen from previous work in the literature. We then tested methodological factors that impact the ability of these graph metrics to predict measures of executive function task performance. Sample size had the largest impact on our models and even successful models accounted for very little sample variance. These results suggest that graph metrics do capture functional network organization associated with executive functions, but that graph metrics alone are too reductionist to capture complex patterns of brain activation associated with executive function behaviors. Study 3 identified and categorized cortical hubs in youths – regions supporting communication between brain networks – and tested their ability to predict executive function task outcomes. We found that cortical hub parcels in youths qualitatively resemble those found in adults, but form unique, youth-specific categories. Additionally, hubs integrating sensorimotor information and executive function networks related to behavioral outcomes. These studies demonstrate how measures of brain network organization and integration can improve our understanding of the developing brain, and their role in supporting goal directed behavior and individual differences.