Abstract

Background:

Subjective memory complaints (SMC) reflect individuals’ self-perceived concerns about cognitive functioning but do not perfectly correspond with objective cognitive performance. This heterogeneity suggests that SMC may capture not only current neuropsychological functioning but also individual differences in cognitive appraisal. Building on evidence that subjective cognitive concerns include trait-like components with genetic influences, we investigated whether genetic liability to neuroticism moderates the association between objective neuropsychological performance and SMC.

 

Methods:

The study included 724 adults aged 55–90 years, including cognitively normal individuals and those with mild cognitive impairment, from phases 1 and 2 of the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE). Neuropsychological performance was assessed using the Korean version of the CERAD battery, with global cognition indexed by the CERAD total score (TS) (z-score) (Han et al., 2014) and memory performance by a composite CERAD memory score (z-score). SMC was measured using the 14-item subjective memory complaint questionnaire (SMCQ) (Youn et al., 2009). Neuroticism polygenic risk scores (PRS) were calculated based on published GWAS summary statistics using a clumping-and-thresholding approach (p < 5 × 10⁻⁸). Associations between objective cognitive performance and SMCQ, and their moderation by neuroticism PRS, were examined using linear regression models with interaction terms.

 

Results:

Objective cognitive measures were significantly associated with SMCQ, indicating that poorer memory and global cognitive performance were related to higher SMCQ scores. Neuroticism PRS showed no main effect on SMCQ, whereas significant interactions between cognitive performance and neuroticism PRS were observed for memory, with a comparable pattern for global cognition (Table 1). In level-specific analyses, associations between memory performance and SMCQ were significant across all neuroticism PRS levels, but the magnitude of the association was strongest at lower PRS levels and progressively attenuated with increasing PRS (Table 2; Figure 1). A similar graded attenuation pattern was observed for global cognition.

 

Conclusion:

Neuroticism polygenic risk moderated the association between objective cognitive performance and SMCQ, with the strongest association observed at lower levels of genetic liability. This suggests that when trait-related neuroticism is low, objective cognitive performance more strongly predicts SMC, whereas higher genetic liability attenuates this relationship.