Longitudinal Within-Person Associations Between Quality of Social Relations, Structure of Social Relations, and Cognitive Functioning in Older Age

Abstract

Objectives: Individuals' social connections and interpersonal experiences can both shape and be shaped by cognitive functioning. This study examines longitudinal within-person associations between quality of social relations, structure of social relations, and cognitive functioning in older age.

Methods: We examined 16-year longitudinal data (3 waves) from 497 older adults (M = 66.07 years, SD = 0.83, range = 64-68 years) from the Interdisciplinary Longitudinal Study of Adult Development and Aging. Quality of social relations was measured by scales on perceived emotional support, instrumental support, and social integration. Structure of social relations was measured by self-reported number of leisure time partner types, indicating social network diversity. Cognitive functioning was assessed as a latent construct consisting of five cognitive tests (i.e., Information, Similarities, Letter Fluency, Picture Completion, Block Design). We used a random intercept cross-lagged panel model in the analysis.

Results: At the within-person level, prior quality of social relations, but not structure of social relations, was positively associated with subsequent cognitive functioning. Moreover, prior cognitive functioning was positively associated with subsequent structure of social relations, but not with quality of social relations.

Discussion: Quality of social relations is a protective factor of cognitive aging. Additionally, responding to prior lower cognitive functioning, social network diversity reduced, but quality of social relations did not seem to change. Overall, this study suggested that social relations and cognitive functioning mutually influence each other, but different aspects of social relations (i.e., quality, structure) might have different directional associations with cognitive functioning.

Keywords: Cognitive aging; Longitudinal; Random intercept cross-lagged panel model; Reciprocal association; Social network diversity.