Should we all go to college? Revisiting educational benefits for memory aging in a large-scale study

Education is a robust predictor of episodic memory, yet its role in memory aging remains ambiguous. Prior studies often restrict analyses to "healthy" subsamples or focus narrowly on impairment, leaving open whether educational gradients simply reflect preserved health.

Using nationally representative Health and Retirement Study data, we followed 15,597 U.S. adults aged 51+ over 16 years (106,173 observations). Dual-outcome latent growth models estimated immediate recall (attention/encoding) and retention (storage/retrieval), conditioning trajectories on time-varying indicators of homeostatic dysregulation (e.g., CRP, blood sugar, BMI), frailty (e.g., grip strength, FEV), mortality, cognitive impairment, and self-reported health and disease.

Educational advantages persisted but did not widen, yielding near-parallel trajectories across groups. College graduates recalled modestly more words at baseline (0.17-0.44 points out of 10) even when health differences were held constant. Gains concentrated in immediate recall, with little evidence for retention or decline-rate differences. Adjusting for adult and parental SES attenuated but did not eliminate these baseline gaps.

College completion conveys small but durable benefits for memory aging, reflecting capacities established earlier in life rather than protection once decline begins. Yet the largest disparities stemmed from health vulnerability and high school non-completion. Interventions ensuring universal high school completion may yield greater public health returns than expanding college access alone.