Reed · Journal of the International Neuropsychological Society : JINS 2011 · Clinicopathological cohort study · n=652

Cognitive activities during adulthood are more important than education in building reserve.

Cited 127 times in the scientific literature.

Level 3 - non-randomized controlled study

Clinicopathological cohort study with postmortem pathology

PubMed 23131600 · doi:10.1017/S1355617711000014 · record verified 2026-08-29

What was done

Analyzed 652 autopsied participants from the Rush Memory and Aging Project and the Religious Orders Study to evaluate which life experiences predict cognitive reserve. Reserve was modeled as a latent variable capturing residual variance in six cognitive domain scores after regressing out neuropathology (Alzheimer's disease markers, Lewy bodies, infarcts, microinfarcts, and brain weight). Predictors in a multivariate regression model included socioeconomic status, years of education, leisure cognitive activities at age 40 (CA40), and leisure cognitive activities at late-life study enrollment (CAbaseline).

What was found

Predictors showed weak to moderate inter-correlations. In the multivariate model, socioeconomic status was not statistically significant. The remaining predictors were statistically significant: CA40 had the strongest positive association with reserve (β = .31), followed by CAbaseline (β = .28), while education had an inverse relationship (β = -.25). Exact p-values, standard errors, and confidence intervals were not reported in the abstract.

Why it matters

These findings suggest that sustained cognitive stimulation throughout adulthood may contribute more robustly to late-life cognitive reserve than early-life formal educational attainment.

Limits

Midlife cognitive activity at age 40 was assessed retrospectively, introducing risk of recall bias. The sample was drawn from self-selected cohorts willing to undergo postmortem autopsy, limiting generalizability to broader populations. The negative coefficient for education in the full model likely reflects statistical suppression or collinearity rather than a true biological harm, and unmeasured confounders were not accounted for.

Cited by