Background: Evidence suggests the importance of physical activity and social engagement in cognitive preservation. Group-based dancing combining exercise and prosocial features may generate physical and cognitive benefits. Objectives: To investigate the association between multiyear habitual square dancing and domain-specific cognitive function, and assess the relative importance and joint impact of physical activity and social activity on cognition. Methods: Using the cross-sectional propensity score matching method, the study compared the mental status, episodic memory, and overall cognitive performances of 145 amateur female square-dancing participants (aged ≥45 y) to their sociodemographic- and health-status–matched 222 nondancing counterparts, selected from the China Health and Retirement Longitudinal Study. Results: The authors found a positive association between multiyear square dancing (average 8 y) and overall cognitive performances (mean difference = 2.84; 95% confidence interval [CI], 1.65 to 4.02), which was apparent in processing capacity (2.29; 95% CI, 1.51 to 3.07) but not in memory (0.55; 95% CI, −0.13 to 1.23). The hypothesized synergic effect of physical activity and social activity on cognition was only observed in group-based exercises embodying these 2 components simultaneously. Conclusions: Long-term square dancing as one type of physically and socially engaging activities may preserve cognition. Future longitudinal and interventional studies are needed to further clarify the causal relationship.