History-graded increases in older adults' levels of cognitive performance across a number of different ability domains are well documented. Less is known, however, about whether such historical advances generalize to measures of verbal learning and whether rates of within-person change therein have also historically shifted. To examine these questions, we used up to 17+-years within-person longitudinal change data obtained in the Manchester Longitudinal Study of Cognition (MLSC). We matched cohorts and weighted analyses by age, number of measurement occasions, and education and compared up to four-wave longitudinal data from 1,279 participants born earlier in historical time (1903-1919, 2,668 observations) versus 1,049 participants born later (1920-1935, 2,239 observations). We applied growth models that orthogonalized between-person age and within-person aging effects, controlled for retest effects, and adjusted for sociodemographic characteristics. We found little to no evidence for historical improvements in verbal learning among older adults. It was only at the zero-order level when participants were in their 60s that later-born cohorts outperformed their matched earlier-born peers. We also found no evidence of historical improvements in the rate of within-person aging-related decline in verbal learning. If anything, the later-born cohort appeared to experience somewhat steeper declines, particularly at older ages. After statistically controlling for sex, occupational status, and city of residence, the only remaining cohort difference was the steeper quadratic (accelerated) rate of decline observed in the later-born cohort. We discuss the theoretical and practical implications of these findings. (PsycInfo Database Record (c) 2025 APA, all rights reserved).