OBJECTIVES To identify differences in cognitive function between elderly adults with sleep apnea syndrome (SAS) and healthy controls. DESIGN Cross-sectional. SETTING Sleep laboratory at Seoul National University Bundang Hospital. PARTICIPANTS Sixty-three elderly adults (26 female; mean age 68.2 ± 4.8) without cognitive disorders. MEASUREMENTS Sleep-laboratory polysomnography findings and cognitive function results determined using the Korean version of the Consortium to Establish a Registry for Alzheimer's Disease Neuropsychological Assessment Battery. RESULTS When the control group (apnea-hypopnea index (AHI) < 15) was compared with the groups with mild to moderate (15 ≤ AHI < 30) and severe (AHI ≥ 30) SAS, significant differences in delayed recall (P = .003) and errors on the Trail-Making Test Part B (TMT B; P = .009) were observed, with participants with severe SAS showing greater impairment on both tests than controls (P = .02 and P = .01, respectively). Stepwise multiple regression showed that oxygen desaturation index (β = -0.37, P = .003) and educational level (β = 0.24, P = .04) determined delayed recall impairment (adjusted coefficient of determination (R (2)) = 17.8%, P = .003). TMT B errors were independently associated with educational level (β = -0.41, P = .001) and AHI (β = 0.31, P = .007; adjusted R (2) = 25.7%, P = .001). CONCLUSION Severe SAS is associated with measures of delayed recall and executive function in cognitively healthy older adults. Although further study is needed, this evidence may provide further rationale for the treatment of SAS in older adults. Moreover, the role of SAS as a risk factor for cognitive disorders needs to be determined.