Cognitive Reserve (CR) capacity can be viewed as the maximum processing potential of neural systems that support adaptive cognitive performance in age-related cognitive decline. CR is a complex construct that cannot be directly measured as it refers to processing efficiency of standard, and non-standard networks. Proxy factors such as psychosocial/lifestyle and cognitive variables are therefore used but are in need of construct validation; and importantly, the cognitive factors potentially involved in CR capacity, such as executive function (EF), may be modifiable. The research objectives were to investigate in healthy adults (a) the validity of an a priori model of CR capacity and cognitive outcomes, and (b) the modifiability of the construct of CR capacity through adaptive EF training. Firstly, the construct validity of Satz et al.’s (2011) four-factor CR capacity model was explored using data from the Maastricht Ageing Study (MAAS) (study 1). Exploratory and confirmatory factor analysis established a two-factor model comprised of executive function/processing resources (EF/PR) and cumulative cognitive enrichment (CCE), which was validated using data from the Irish Longitudinal Study on Ageing (TILDA) (study 2). Predictive relationships between this model and global cognition/memory outcomes were also explored (study 3) and replicated (study 4), providing support for a strong, positive predictive relationship between CR capacity and outcomes at baseline and follow-up. Healthy older adults with variable memory concerns (aged ≥50) were profiled to further investigate relationships between the CR capacity model parameters and global cognition/memory (study 5). EF/PR was again found to be predictive of global cognition/memory scores. The effects of a targeted novel response inhibition training intervention (active control vs. experimental) aimed at boosting a formative component of CR capacity (EF/PR) were investigated (study 6). Both levels of training were found to have direct and near transfer effects following five weeks of training. However, effects did not generalise to far transfer measures (e.g., global cognition/memory outcomes). Overall, this novel approach to modelling CR suggests that control processes are an important contributor to CR capacity, are modifiable, and therefore represent a promising target for future interventions aimed at improving cognitive function in healthy ageing.