BACKGROUND: A new model of depression is proposed which is biological and explanatory on the behavioral level. It is hypothesized that a patient suffering from a depressive disorder loses the ability to produce one or more modes of behavior at a given time and simultaneously has the urge to produce one or more other modes of behavior constantly. The patient is also unable to interpret his behavioral disorder ("loss of self-understanding"). MATERIAL/METHODS: The authors developed the Salzburg Subjective Behavioral Analysis (SSBA) self-assessment questionnaire asking subjects to elaborate on changes in 35 modes of behavior and to interpret the behavioral disorder, if possible. The study was conducted using this questionnaire and the Hamilton Depression Scale on 30 inpatients with depressive disorders and this questionnaire on 30 healthy controls. RESULTS: Extreme positions in the SSBA were found in all 30 patients, compared with 2 in the 30 healthy controls. Significant correlation between the occurrence of extreme positions and the Hamilton scores was found (r=0.44, p=0.02) and correlation between the Hamilton scores and the inability of the patients to interpret the behavioral disorder was nearly significant (r=-0.35, p=0.06). CONCLUSIONS: The occurrence of extreme positions in the SSBA increased with depressive mood. In addition, the likelihood that subjects could explain their extreme positions in behavior decreased with increasing depression. The SSBA questionnaire assesses the self-evaluation of modes of behavior. As a diagnostic instrument, DSM-IV does not have criteria for modes of behavior. Thus, behavioral analysis should be considered in diagnosing depression.