In the context of increasing concern in occidental countries about dementing diseases in the geriatric population, an epidemiological study has been done in the city of Zaragoza, Spain. The sample, randomly selected from the census, was stratified by age and sex and included 1,134 elderly (aged 65 plus years) living in the community. In phase I, lay interviewers administered the Spanish versions of the Geriatric Mental State (GMS) and Mini-Mental Status Examination to the elderly. The individuals were considered to be "probable cases" on the basis of GMS "global" scores previously reported to be valid. These "probable cases" were then examined in phase II by standardized psychiatrists using both, the GMS and MEC, but also the History and Aetiology Schedule (HAS). A proportion of "probable normals" were also examined by the psychiatrists to assess the validity of the screening procedure in the community. The identified "cases" of dementia and "cases" of depression were then studied by the neurologists in what we call phase III. Specific project criteria for dementia were used but, in fact, all identified "cases" of dementia fulfilled DSM-III criteria. Some data of the application of the AGECAT computer program in both, phase I and phase II are now reported. The different severity levels of cognitive deficits according to AGECAT criteria are compared with clinical criteria. The advantages and disadvantages of the computer program in the detection of minor cognitive difficulties are discussed.