This thesis explores the lifestyle and brain structure correlates of cognitive ageing. Using large datasets and a range of multivariate statistical approaches, it is divided into three empirical sections, which are framed by a General Introduction and Conclusion.
The first Chapter, the General Introduction, discusses this thesis within its wider context: ageing populations pose significant issues for individuals and societies, while the rise of Big Data offers promising opportunities to study the effects of brain health and lifestyles on cognitive ageing. Chapter 2 investigates the relationship between lifestyle and cognitive abilities across the adult lifespan. It shows that, in a sample of cognitively healthy participants (Cam-CAN; N=708), higher education, better physical and mental health, more social engagement and a greater degree of intellectual engagement are each correlated with better fluid and crystallized cognitive abilities. Chapter 3 focuses on brain structure. It explores how different aspects of morphology – such as cortical thickness, curvature, sulcal depth or surface area – change with age, and relate to cognitive outcomes, including fluid intelligence. This chapter’s main finding is a cross-sectional and longitudinal double dissociation: while cortical thickness declines rapidly with age, it does not relate strongly to cognition, particularly after adjusting cognition for age. Surface area, on the other hand, has only moderate age-effects, but captures cognitive difference and change well. These findings replicated in a second, independent dataset (LCBC; N = 1236), suggesting that they are robust and generalizable across cohorts. It is plausible that this hitherto largely overlooked dissociation reflects two distinct neural features, which I discuss at the end of this section.
The final empirical section, Chapter 4, brings together the first two sections by assessing lifestyle, brain structure and cognition simultaneously. Specifically, it investigates whether brain structure mediates the relationship between lifestyle and cognitive abilities. After initial cross-sectional analyses in Cam-CAN using the lifestyle factors created in Chapter 1, it then focuses on the mechanistically more plausible and (because of its longitudinal nature) statistically more robust relationship between cardiovascular health and cognitive performance in the LCBC data. Finally, in the conclusion, I address the wider societal and policy context in which this PhD is, ultimately, embedded. This thesis – and indeed the science of healthy minds and brains in general – finds its urgency not least in the steadily increasing speed at which the older segment of many nations’ populations is growing in size and proportion. Understanding (healthy) ageing amidst this demographic shift is of vital importance to individuals, society, and governments alike. I therefore reflect on my experiences at the World Health Organization (where an internship enabled me to contribute to the WHO’s report on Ageism) and on the kind and quality of science which should (and, perhaps, should not) inform policies designed to ensure that living a longer life also means living a healthier one.