Cognitive activity, cognitive function, and structural MRI findings

Abstract Background More frequent engagement in late‐life cognitive activities (e.g. reading, playing games) has been associated with slower cognitive decline in old age. Even after adjusting for neuropathological burden, cognitive activities were associated with less cognitive decline. However, it is not known whether people who engage more frequently in cognitive activities show differences in brain anatomy compared to those who participate less frequently. We tested the hypothesis that brain anatomy, as measured by structural MRI indices, mediates the association of cognitive activities and cognitive decline in old age. Method In a sample of 975 participants (60% African American and 60% female) from the Chicago Health and Aging Project, late‐life cognitive activities were collected at baseline; MRI was conducted an average of 7.9 years later. We identified several significant correlations of late‐life cognitive activity with MRI indices (brain regions and cortical thickness) and then used linear mixed‐effects regression models to examine those associations with global cognition. Global cognition was assessed using a standardized measure of episodic memory, perceptual speed, and the MMSE, before and after MRI scans. Result Higher frequency of overall cognitive activity was associated with seven brain volumes and two cortical thicknesses, which were associated with global cognitive level. Of those associations, five were associated with cognitive decline and explained about 42% of the cognitive activity association with cognitive decline. Specifically, a higher reading frequency was associated with lower white matter hyperintensity volume (WMH) and third ventricular volume. Higher WMH and third ventricle volumes were also associated with lower global cognition and a faster rate of cognitive decline. We also found that a higher frequency of playing games was positively associated with greater hippocampal and parahippocampal volumes and caudal anterior cingulate thickness in the left hemisphere. Conclusion Individuals who engage more frequently in cognitively stimulating activities show differences in brain anatomy compared to those who participate less frequently, as measured by structural MRI. These results suggest that cognitive activities are related to improved structure and function of the brain, thereby shedding light on the well‐established findings that cognitive activities are associated with cognitive decline independent of neuropathologic burden.