Potential biases due to acute depressive symptomatology on raters' assessments of social maladjustment derived from patients' reports were assessed in 25 patients responding to pharmacotherapy during medium-term hospitalization. Patients were questioned on two separate occasions about their social maladjustment covering the exact same period (the 4 months preceding hospitalization): the first was during the acute illness phase, and the second a mean of 20.5 days later, when symptoms remitted. In the second report, composite scores for all fields as a whole showed significantly fewer reports of social impairment than did the first. Significant differences from the first to the second evaluation concerning both subjective distress and observable behavior were found in four and three, respectively, of the five "fields" of social adjustment. Although subjective distress was most modified by remission of acute symptoms, even supposedly objective, observable disturbances were significantly affected. These results indicate that acutely depressed patients overreport social maladjustment, which they then more accurately reappraise when symptoms remit. Patients are completely unaware of both the initial bias and of the reappraisal.