TOPOLOGICAL AND COMPUTATIONAL MECHANISMS OF RESILIENCE IN THE AGING HUMAN BRAIN

Aging is associated with declines in cognitive function. Declining cognition can impact quality of life and also pose a broader societal public health concern. However, there is significant variation in cognitive decline across individuals, with some individuals exhibiting cognitive resilience, allowing them to maintain high levels of cognitive performance into older adulthood. Understanding the mechanisms supporting cognitive resilience could enable us to develop strategies that mitigate age-related declines in cognitive function, and improve the quality of life for older individuals. In this dissertation, we investigated network resilience, and redundancy, as mechanisms that could promote cognitive resilience in healthy-aging.By simulating perturbations that mimic age-associated changes in network connectivity, we can assess the resilience of an individual’s functional brain networks to these processes. In the second chapter, we improve upon existing methodology for studying network resilience in brain networks by applying k-shell decomposition to delineate the core structure and core resilience of functional brain networks in middle and old-aged adults, and assess the relationships between these features and cognitive function. We observed positive relationships between episodic memory and robust core connectivity, and core resilience. Additionally, we discovered that overall core resilience mediated differences in age-related changes in cognition, suggesting that core resilience could protect against cognitive decline. In the third and fourth chapters, we expanded up existing work that investigated the neuroprotective role of redundancy in cognitive-aging. Redundancy, defined as the existence of duplicate pathways between brain regions, is believed to support cognitive resilience in healthy and pathological-aging by enabling robust communication between brain regions even if some pathways fail due to age or pathology-associated changes. In the third chapter, we discovered that redundancy mitigates the impact of age-associated network changes on network controllability, and has a stronger relationship with processing speed than grey matter volume, a traditional measure of brain reserve, and network controllability. The final chapter found that redundancy protected the processing speed abilities from negative impacts of accelerated brain aging in healthy-aging. This dissertation presented initial evidence for core resilience as a mechanism supporting cognitive resilience in healthy-aging, and provided further support for the hypothesized neuroprotective role of redundancy in brain networks.