PURPOSE: Decreased physical and cognitive ability have become seriously problematic in aging. Studies seldom describe the relationship between physical activity (PA), physical performance and cognitive abilities, most of which by measuring only one or two specific abilities using questionnaires. The study was thus designed to examine the relationship between PA and physical performance and multiple cognitive abilities in the Chinese elderly. METHODS: The design was a cross-sectional study. 148 people aged 65.15±7.79 were included from Guangxi Province in southern China. PA was measured by the Physical Activity Scale for the Elderly (PASE, Cronbach’s ɑ=0.752; Test-retest reliability=0.980). Physical performance included coordination and lower limb strength, functional walking ability and the balance ability. Cognitive abilities were measured by computer-based tests, including simple reaction time and executive function (stroop task and shifting task). The association between PA and the outcomes were examined using correlation matrix and linear regression. RESULTS: 1) The average time spent on physical activity was 1.8h/d; 2) High levels of PA significantly associated with better gait speed (B=-0.67, p<0.05), 30-s chair-stand test (B=0.72, p<0.05) and reaction abilities (B=0.89, p<0.05); 3) The gait speed significantly correlated with the accuracy of stroop task (r = 0.23, p = .018) and shifting task (r = 0.21, p = 0.023), grip strength significantly correlated with the accuracy and the reaction time of the stroop task(r = 0.19, p = 0.04 & r = 0.19, p = 0.04). However, other physical performance outcomes had no significant correlation with the cognitive abilities. CONCLUSIONS: More physically active residents scored higher on physical performance outcomes. Positive associations were found between physical performance, executive function and reaction ability.