Bilingualism and "brain reserve" in subregions of the hippocampal formation

Abstract

With aging, the hippocampal formation shows variable structural atrophy, which is associated with a decline in cognitive performance. Bilingualism is related to higher hippocampal gray matter volume (GMV), potentially representing a form of brain reserve in aging. However, the differential influence of bilingualism on hippocampal subregions remains unclear. Thus, we investigated GMV differences and differences in age-GMV relationships between mono- and bilinguals in the hippocampal formation and its subregions, hippocampus proper and subicular complex. We included 661 adults aged 19 to 85 years (257 monolinguals, 404 sequential bilinguals, predominantly native German speakers with variable second language background) from the population-based 1000BRAINS cohort. GMV differences in mono- vs. bilinguals were assessed for six regions of interest (hippocampal formation, hippocampus proper, and subicular complex; each left and right) using analyses of covariance. Effects of bilingualism on age-GMV relationships were investigated via moderation analyses. We found higher GMV in bilinguals in the bilateral subicular complex, while only a trend towards this effect existed for the hippocampal formation. Moderation analyses revealed similar age-GMV relationships between mono- and bilinguals for all regions of interest. Higher GMV in bilinguals' hippocampal formation seems specifically attributable to the subicular complex rather than the hippocampus proper. With similar age-GMV relationships for mono- and bilinguals, bilingual brain reserve in the subicular complex may persist over time. This may be particularly beneficial since subicular atrophy has previously been associated with higher risk for dementia. Altogether, a differential impact of bilingualism on hippocampal subregions has been demonstrated.

Keywords: Aging; Bilingualism; Brain reserve; Gray matter volume; Hippocampus; Subicular complex.

Fig. 1

Subregions of the hippocampal formation based on the cytoarchitectonic probabilistic Jülich-Brain atlas [75] overlayed on an MNI template on a rendered brain (a) and in coronal view (b). Dentate gyrus (DG) and cornu ammonis (CA; including CA1, CA2, and CA3) formed the hippocampus proper (see c); the subicular complex consisted of prosubiculum (ProS), subiculum (Sub), presubiculum (PreS), and parasubiculum (PaS). R, right; S, superior

Fig. 2

Differences in GMV between mono- and bilinguals of the total sample for the bilateral hippocampal formation (a), hippocampus proper (b), and subicular complex (c). Taking into account the best-fitting model for the relationship between GMV and age within the respective ROI, analyses for the bilateral hippocampal formation and hippocampus proper were based on a quadratic relationship between age and GMV, while analyses for the bilateral subicular complex were based on a linear relationship between age and GMV. Boxplots show median values, lower and upper quartile, maximum and minimum values within the 1.5 interquartile range below/above the lower/upper quartile, and outliers within each language group (please note: these are not outliers with respect to the complete data set). GMV, gray matter volume

Fig. 3

Age-GMV relationships in mono- and bilinguals for the left (a) and right hippocampal formation (b), left (c) and right hippocampus proper (d), and left (e) and right subicular complex (f). Regression lines are depicted with 95% confidence intervals (hippocampal formation and hippocampus proper: quadratic model for the relationship between age and GMV; subicular complex: linear model for the relationship between age and GMV; separately for mono- and bilinguals). The gray underlay indicates participants ≥ 55 years of age, who were included in the older subsample. GMV, gray matter volume