Human brain aging is a multifactorial process that involves environmental, genetic, and behavioral factors. A typical aging process often presents substantial individual heterogeneity. Some individuals experience a faster pathological aging process than do others, whereas others maintain a longer healthy brain function. The individual variations observed in these adults suggest that some people are more resilient to pathological aging than are others; large brain size and high educational attainment may contribute to improved tolerance for deterioration changes with aging. (Bartrés-Faz and Arenaza-Urquijo, 2011, Satz, 1993, Fratiglioni and Wang, 2007, Stern et al., 2019, Stern, 2012, Stern, 2002, Cox et al., 2016) Several studies have utilized the brain reserve (BR) hypothesis employing intracranial volume (ICV), cognitive reserve (CR) hypothesis employing education level, and brain resilience hypothesis to explain individual differences during aging. (Bartrés-Faz and Arenaza-Urquijo, 2011, Satz, 1993, Fratiglioni and Wang, 2007, Stern et al., 2019, Stern, 2012, Stern, 2002, Hohman et al., 2016) These hypotheses suggest that greater intrinsic or extrinsic capacity of an adult brain may be beneficial in dealing with brain pathology to minimize symptomatology. (Bartrés-Faz and Arenaza-Urquijo, 2011, Satz, 1993, Fratiglioni and Wang, 2007, Stern et al., 2019, Stern, 2012, Stern, 2002, Hohman et al., 2016)

Large brain size and higher educational attainment at an earlier age could have beneficial effects on regional brain size while aging. (Van Loenhoud et al., 2018, Roe et al., 2011, Kim et al., 2015, Mitchell et al., 2020, Cox et al., 2016) ICV has routinely been used in brain aging research (Van Loenhoud et al., 2018, Roe et al., 2011, Han et al., 2020) as it is positively associated with one's subregional brain volumes and intelligence. (Mitchell et al., 2020) Several studies have reported a positive correlation between intelligence and ICV. (Mitchell et al., 2020, Nave et al., 2019, Pietschnig et al., 2015) A large ICV could be protective against clinical manifestations in later life. (Van Loenhoud et al., 2018) Higher education could also have a protective effect against cortical thinning in cognitively normal (CN) older individuals (Kim et al., 2015) and may be positively related to brain volume. (Walhovd et al., 2020)

Concurrent investigation of the factors affecting brain aging is important because factors determined in early life, such as educational attainment or ICV, may also be associated with brain aging in later life, such as with regional brain volume. However, the current literature generally reports ICV and educational attainment as nuisance variables. The investigation of ICV is important because ICV in later life roughly represents the maximal brain size of individuals. (Narvacan et al., 2017) Since most brain aging studies generally focus on older participants who have already undergone extensive changes, mostly atrophy of the brain since achieving its maximal size, ICV may provide a good substitute for full-grown brain size. ICV could also reflect the resilience effect in later life, where education may play an important role in determining full brain growth in early life. (Narvacan et al., 2017) Very few studies have simultaneously examined brain size and education level in association with aging, reporting that the time to incident cognitive impairment is moderated by education and brain volume. (Roe et al., 2011)

Herein, we aimed to understand better how ICV and educational attainment, which could serve as surrogates for individual resilience capacity, jointly influence age-related properties in subregional brain volumes in men and women. Specifically, we investigated whether the effect of educational attainment on brain volume is mediated by ICV later in life.