Affective symptoms, such as depression and anxiety, are commonly comorbid with dementia. Estimates report that up to 40% of people with dementia also present with symptoms of depression. Research has also shown that a longitudinal association between affective symptoms and cognitive function may exist over time. However, the majority of research has focussed on older samples (aged 60+ at baseline) with short periods of follow up (<10 years). Less is known about how the association between affective symptoms and cognitive function operates over a longer period of time, including much earlier in the life course. The studies within this thesis aim to test the nature of this complex association, using a life course approach. The first study in this thesis used a systematic review and meta-analysis method and concluded that depression in older adulthood was associated with a faster subsequent cognitive decline. The second study used secondary data from the National Survey of Health and Development (NSHD), a nationally representative birth cohort of over 5,000 people born in England, Scotland and Wales, during one week of 1946. This study found that adolescent, but not adult, affective symptoms significantly predicted poorer baseline memory and information processing speed scores, but not rate of decline over time. The third study also used data from NSHD to test bidirectionality between affective symptoms and cognitive function. This study found that affective symptoms significantly predicted subsequent level of memory and processing speed over a 16 year period, but that this association did not operate in the opposite direction. The fourth study used data from the National Child Development Study (NCDS), a population representative birth cohort of over 18,000 people born in England, Scotland and Wales, during one week of 1958. This study revealed that accumulation of persistent affective symptoms was a better predictor of midlife memory function than sensitive periods. The fifth and final study in the thesis also used data from NCDS to explore inflammation as a potential mediator in the association between affective symptoms and cognitive function. This study showed that C-Reactive protein (CRP) fully explained the association between adult affective symptoms and memory function, suggesting that inflammation is an important explanatory factor of this relationship. Taken together, these findings suggest that chronic and persistent affective symptoms across the life course play an important but complex role in contributing to cognitive function across mid to later life, with particularly strong effects observed on memory function. In addition, inflammation may be one important explanatory mediator of the association between affective symptoms and midlife memory function.