AbstractBackgroundPrevious studies on the effects of intergenerational social mobility on cognitive outcomes were not consistent, additionally, underlining brain anatomical variation is unclear. In the present study, we aim to examine the associations of social mobility with brain structure and cognitive trajectories in older adults.MethodThe study included 771 participants, ages 69.8 ±5.2 years, of the Whitehall II ‐ MRI substudy (2012‐2016) with data on their father’s occupation, their own occupation in mid‐life, lifestyle (smoking, drinking, physical activity, social activity, sleep disturbances, auditory difficulty, depression, obesity, hypertension), neuroimaging (T1 structural MRI) and cognitive abilities (global cognition, crystallized cognitive ability, and fluid intelligence). Social mobility was classified into stable high social class, stable low social class, upward mobility and downward mobility group based on occupational shifts from father. We used the ANCOVA tests to assess the association of social mobility with gray matter volume (GMV) and cortical thickness (CT) in various brain regions, and the growth curve modeling (GCM) to evaluate its relations to cognition over 20 years, adjusting for social lifestyle covariates and associated GMV/CT regions identified above.ResultSocial mobility was significantly related to 48 out of the 136 regions of GMV across the frontal lobe, limbic lobe, and temporal lobe; and to 4 out of the 68 regions of CT at the whole‐brain level. Stable high social class preserved the largest GMV and CT value; pair‐wise results suggest upward mobility preserved a larger GMV value than the downward mobility group (average diff. in cm3: 0.121, P<0.05); and similarly a larger CT value (average diff. in cm:0.029, P = 0.051). In GCMs, stable high social class (ref. downward mobility) had a higher entry score of fluid intelligence (b = 2.617, P<0.05), all other comparisons including the entry score and decline rate of cognitive abilities were non‐significant.ConclusionOur results suggest a lack of direct association between social mobility and trajectories of global cognition and crystallized cognitive ability; instead, an indirect effect through preserving the value of brain GMV and CT may occur. Additionally, an adverse causal relationship between social mobility and fluid intelligence might happen.