Background: Cross-sectional epidemiologic studies support that high salt intake increases the risk of cardiovascular disease, but there are little data on the relationship between salt intake and left ventricular (LV) function in the general population. Methods: We analyzed cross-sectional data of 1,228 participants, who were aged 45-74 years and free of known cardiovascular disease in 2007, from one of population-based cohorts of the Korean Genome Epidemiology Study (KoGES). All individuals completed a salt usage behavior questionnaire including a question on preference for salt taste, which was reportedly associated with 24-hour urinary sodium excretion, and underwent two-dimensional echocardiography and tissue Doppler imaging (TDI) at mitral annulus for assessment of LV function. Results: After taking into account potential risk factors including body mass index, measures of glucose and lipid metabolism, and blood pressure, the preference for salt taste was inversely associated with peak mitral annular early diastolic velocity; low (reference = 7.20 cm/s), moderate (7.12 cm/s), and high preference for salt taste (6.80 cm/s) (P value, test for trend <0.001). In analysis stratified by sex or body mass index, the association between high preference for salt taste and reduced LV diastolic function was stronger among men (P value <0.01) or overweight persons (P value <0.05) rather than among women or those with normal weight. Conclusion: In our large population-based sample, the preference for high-salt taste was strongly and independently associated with impaired LV diastolic function, particularly among men or overweight persons.