OBJECTIVE: Ecological tests are useful in assessing executive function deficits and may be of value in appraising response to treatment in Alzheimer's disease patients. Our aims were to examine executive function using the Behavioral Assessment of the Dysexecutive Syndrome for a Brazilian sample of older-adult controls and probable early Alzheimer's disease patients, and verify the applicability of this test battery. METHOD: Forty-one older-adult controls were matched with mild Alzheimer's disease patients by age, education, and gender. RESULTS: There significant inter-group differences in overall profile and almost all subtests except temporal judgment, time spent on planning the first and second Zoo Map visit, number of errors when copying drawings, naming pictures and Six Modified Elements arithmetic, and dysexecutive questionnaire self-rating. The Behavioral Assessment of the Dysexecutive Syndrome item that best discriminated controls from patients was the Modified Six Elements - adapted (general index), with a sensitivity of 80% and specificity of 90%, (AUC = 0.91, p