To examine whether poor objective and subjective sleep quality are differentially associated with cognitive function. Cross-sectional. Participants were recruited from primary and secondary care, and directly from the community, in Sydney, Australia. The sample consisted of 74 men 50years and older (mean [SD], 58.4 [6.2] years), with comorbid depression and above-threshold insomnia symptoms, participating in a trial of online cognitive behavioral therapy for insomnia. Insomnia severity and depression severity were assessed via self-report. Objective sleep efficiency and duration were measured using actigraphy. Objective cognitive function was measured using 3 subtests of a computerized neuropsychological battery. Poor objective sleep efficiency was associated with slower reaction time (r=-0.249, P=.033) and poorer executive functioning (odds ratio, 4.14; 95% confidence interval, 1.35-12.69), but not memory. These associations remained after adjusting for age, education, depression severity, cardiovascular risk, and medication. Subjective sleep quality was not related to cognitive function. Among older men with depression and insomnia, objectively measured poor sleep efficiency may be associated with worse cognitive function, independent of depression severity. Objective poor sleep may be underpinned by neurobiological correlates distinct from those underlying subjective poor sleep and depression, and represent a potentially effective modifiable mechanism in interventions to improve cognitive functioning in this population. This supports the use of objective measures of sleep in diagnostic assessments and care.