Dementia-related neuropathological changes begin decades before clinical symptoms emerge, making mid-life a critical window for prevention. Females have a higher incidence of Alzheimer's disease (AD) than males, even after accounting for longer lifespans, and would therefore particularly benefit from dementia prevention strategies. Building cognitive reserve has therefore become a key preventative strategy, yet it remains unclear whether reserve contributors operate differently in females and males during mid-life, and how dementia risks interact with these pathways. To address these questions, I leveraged the PREVENT-Dementia program, a large multi-site cohort of cognitively healthy mid-life adults at risk for late-life dementia (N = 700; 40-59 years), to examine sex differences in the associations between reserve contributors, dementia risks, cognition, and the brain.

Chapter 2 investigated whether reserve contributors enhance cognition in mid-life, and whether these effects are moderated by biological sex and APOE4 status. I found that sex and APOE4 interacted in moderating the associations between occupational attainment and episodic and relational memory. APOE4 carrier females with higher occupational attainment showed better cognitive performance, whereas APOE4 carrier males showed poorer performance.

Chapter 3 examined whether education moderates sex differences in the associations between stimulating activities, cognition, and brain structure. Results showed that males with lower education exhibited positive associations between engagement in stimulating activities and multisensory processing, as well as greater global cortical thickness, particularly among those with higher cardiovascular dementia risk. No comparable education-dependent effects were observed in females.

Chapter 4 focused on sex differences in brain structure-cognition relationships in mid-life. I found significant sex-specific associations, such that greater global cortical thickness was associated with better episodic and relational memory in males, but not in females. These effects were primarily driven by the precuneus region.

Chapter 5 extended this work to brain function using connectome-based predictive modelling. I found that resting-state functional connectivity significantly predicted episodic and relational memory in mid-life, with distinct predictive networks in females and males: females relied more heavily on default mode network connectivity, whereas males did not show this significant relationship. Higher cardiovascular dementia risk was associated with weaker, whereas greater engagement in stimulating activities were associated with stronger memory-related functional connectivity.

Overall, this thesis suggests that building cognitive reserve is not a one-size-fits-all phenomenon, but rather reflects sex-specific pathways and their interaction with genetic and cardiovascular dementia risks. These findings highlight the need for tailored, sex-informed dementia prevention strategies and for promoting equitable access to cognitively stimulating opportunities during mid-life.