Sex Differences in Late-life Cognition: A Psychometric Network Analysis of Well-functioning Older Adults.

Objective: Unidentified sex differences in old-age cognition may emerge in psychometric networks, which look beyond mean scores into the unique cognitive structure of males and females. Accordingly, this study aims to examine cognition in well-functioning older males and females with psychometric network analysis.

Methods: The current cohort (N = 2,802) of community-dwelling adults (≥65 years) was derived from the Advanced Cognitive Training for Independent and Vital Elderly study. Late-life cognition was assessed with the Mini-Mental State Examination. Psychometric networks were produced according to standard guidelines. Analyses were conducted separately for males and females and, as complementary analyses, for males and females of different age groups.

Results: In this sample (N = 2801; Females: N = 2125, 75.87 %; Males: N = 676, 24.13%), late-life cognition clustered into a single subnetwork structure with language-naming a core cognitive process, regardless of sex. The female network had fewer negative connections (N = 1) than the male network (N = 9). Complementary analyses of older ages (≥73 years) showed similar results, whereas the younger (<73 years) female network had more negative connections (N = 4) than the same-age male network (N = 0).

Conclusions: Certain aspects of cognition, like the general structure and core processes, were similar for both sexes. Other aspects, including negative connections between cognitive processes, possibly representing neuro-cognitive compensation, were sex-specific, implying sex differences in cognitive aging patterns (i.e.,males may compensate for age-related neural decline more than females, while females may compensate earlier than males). Practitioners may, therefore, wish to pay greater attention to connections between cognitive processes rather than standard performance metrics to better understand late-life cognition.