Rapidly expanding interest in neuropsychological assessment in Greece has made the development of appropriate culture-specific normative data for core neuropsychological measures essential. In the present study, we sought to establish normative, test-retest reliability and discriminant validity data for the Ruff 2 and 7 Selective Attention Test in the Greek adult population. We administered the test using standard procedures to 218 healthy Greek adults (95 men), aged 17-80 years and two adult patient groups (26 detoxified opiate addicts and 23 HIV seropositive individuals). Using linear regression analyses, we examined the contribution of age, education and gender on Ruff 2 and 7 performance. We further examined test-retest reliability by administering the test on two occasions to 40 healthy adults, with an intersession interval of 12-14 weeks. The regression analyses revealed that age and education, but not gender, contributed significantly to participants performance, with older age and lower education contributing to poorer performance on Speed scores, but only education contributing moderately to Automatic Detection Accuracy scores. Test-retest reliability was very high (.94-.98) for Speed scores, and adequate to high (.73-.89) for Accuracy scores. Younger adults also demonstrated larger practice effects compared to older participants. The test appears to discriminate adequately between the performance of detoxified opiate addicts and HIV seropositive patients and matched healthy controls, as both patient groups performed more poorly than their respective control group. We present normative data for Speed and Accuracy scores stratified by age and education for the Greek adult population.