Recent epidemiological and family genetic studies in different countries using standardized diagnostic interviews for mental disorders have rather consistently demonstrated considerably lower current (e.g. ECA Study: 0.9%) and lifetime (1.4%) prevalence estimates of Major Depression in the elderly (older than 65 years of age) as compared to younger age groups (e.g. 30-44 years: 1 year, 3.9%; lifetime, 7.5%). Some investigators have questioned the validity of these data and suggested alternative interpretations. One possibility is that the complex standardized symptoms and clinical probe questions, and the required judgmental process inherent in diagnostic interviews exceed the cognitive capacity of older adults. This may result in systematic response bias. This paper examines the degree to which the lower prevalence estimates of depression in the elderly are biased due to specific characteristics of the assessment strategy. Analyses of epidemiologic data from the Munich Follow-up Study (MFS), based on the Diagnostic Interview Schedule, demonstrate that (1) older respondents report lifetime depressive symptoms with the same frequency as younger respondents. The additional probe questions designed to identify the degree to which symptoms were caused by factors other than psychological revealed that (2) the elderly more often attribute such symptoms to physical illnesses or conditions. This results in (3) the exclusion of the reported symptoms as a basis for diagnosing depression. A laboratory study demonstrated that "working memory capacity" was a good predictor of this response behavior, indicating that the complexity of the formalized questions exceeds the cognitive capacity of the elderly. Attributing symptoms to a physical illness or condition might be a heuristic strategy to simplify complex recall and judgment processes; the resulting answer is plausible but incorrect. We recommend that the symptom and probe questions of standardized diagnostic interviews be simplified, especially for use with the elderly.