White matter lesions (WMLs) are a prevalent phenomenon in aging that have been reported to impact cognitive functioning. While the scientific literature has highlighted the impact of WMLs on specific cognitive domains, such as memory, processing speed, attention, and executive function, there remains ambiguity regarding if lesions more significantly impact one domain over the other. Further, it is thought that examining lesions based on their location in relation to the lateral ventricles, instead of examining total lesion volume, could be informative in understanding cognitive aging. More specifically, periventricular WMLs (PWML) are more closely associated with cognition and deep WMLs (DWML) are more closely associated with mood disorders (Kim et al., 2009). Additionally, different lesion types are thought to have different etiologies. However, these findings have been inconsistent and more research is needed to solidify this distinction. Considering these gaps in knowledge, we aimed to evaluate the impact of WMLs on short-term memory, long-term memory, and processing speed to explore if one domain is impacted more than the other. Additionally, we explored if certain lesion types (PWML and DWML) predict performance on specific cognitive domains.

The study included 118 cognitively normal older adults aged 55-80 years. We assessed short-term verbal memory as the number of words correctly recalled on all five immediate recall trials on the Rey Auditory Verbal Learning Test (RAVLT); assessed long-term verbal memory as the number of words correctly recalled on the delayed recall trial on the RAVLT; and quantified processing speed as the number of accurately completed items on the Digit Symbol Substitution Test (DSST). Correlations as well as linear regression models were run to investigate WML’s impact on short-term memory, long-term memory, and processing speed in separate models. Age, sex, and education were covariates.

Although total WML volume, PWMLs, and DWMLs were significantly correlated with performance on the RAVLT immediate and delayed recall, the relationships become non-significant when controlling for age, sex, and education. In contrast, all three lesion volumes were correlated with performance on the DSST and remained significant predictors with even when covarying for age, sex and education. Greater age is associated with worse DSST performance, more years of education is associated with better performance on the DSST, and lower total WML, PWML, and DWML volumes were associated with better DSST performance.

Our results show that WMLs are more strongly related to slower processing speed than short and long-term memory. Further, both PWML and DWMLs are important predictors of performance on the DSST but not on memory tasks. Additionally, different types of memory are not impacted differently by lesion type. Further research is necessary to understand if lesion type predicts performance differently in other cognitive domains.