Cognitive reserve in figural and verbal memory in healthy older adults – preliminary results

Abstract

AbstractBackgroundCognitive reserve plays an important role in individual trajectories of age‐related memory decline, with education and semantic fluency representing cognitive reserve proxies (CRPs). Higher CR may attenuate well‐known effects of ageing on the brain, such as decreases in grey matter volume (GMV). We hypothesised that the known associations between age‐related GMV and cognition would be (differentially) attenuated by CRPs.MethodWe analysed 3‐Tesla T1‐weighted Multi‐Echo MPRAGE and cognitive data (Verbal Learning and Memory Test (VLMT), Rey Complex Figure Test (RCFT), and Regensburg Verbal Fluency Test (RWT)) from 51 cognitively normal adults (73.2±7.62 years, 31 male) in our new cohort established within the CRC1436 (https://sfb1436.de/). MRI data were pre‐processed and analysed using CAT12. We performed a whole‐brain linear regression to first assess age‐related reductions in GMV (covariates: sex, total intracranial volume (TIV), education), extracting mean betas for significant regions of interest (ROIs). We investigated the relationship between GMV in age‐vulnerable regions and memory using linear regressions in R (covariates: age, sex, TIV), and assessed CRPs as moderators on this relationship. Finally, we tested whether CRPs explained variance in memory performance over and above that of GMV and age. We accounted for multiple testing via Benjamini‐Hochberg‐correction.ResultWhole brain analyses identified bilateral hippocampal GMV as negatively associated with age (Fig.1). Linear regression analyses yielded a significant positive association between bilateral hippocampal GMV and figural memory (Fig.2). However, we observed no significant moderation by CRPs on this relationship. Finally, we observed weak evidence for cognitive reserve when adjusting for GMV and age, with higher variance explained by CRPs (Table 1). Semantic fluency positively predicted a figural memory composite when adjusting for right (p = 0.006) as well as left hippocampal GMV (p = 0.014). Education positively predicted verbal memory when adjusting for right (p<0.001) as well as left hippocampal GMV (p<0.001).ConclusionConsistent with operational definitions, our results provide weak evidence for semantic fluency and for education as CRPs in figural and verbal memory, respectively. However, our findings are preliminary, given the future addition of more participants. Nevertheless, these results provide important first insights into cognitive reserve in this developing cohort.