This study examines the association between cortisol secretion and fear perception in remitted patients to identify mechanisms underlying risk for recurrence of depression. We hypothesized that the stronger the association between cortisol secretion and fear perception in persons with remitted depression, the more recurrence would be experienced. We also investigated whether high levels of cortisol and fear perception per se predict more recurrence. These effects were assumed to be stronger in women than in men. In a prospective design, we investigated 77 outpatients with remitted depression and related the association between their 24-hour urinary free cortisol secretion and fear perception (from ambiguous faces and from vocal expressions) to recurrence of depression within 2 years. We applied Cox regression models, partial correlations, and Fisher z tests. In 21 patients, depression recurred. Irrespective the channel of perception (eye or ear), the interaction between fear perception and cortisol secretion was significantly related to recurrence of depression. Patients high or low on both variables are more at risk. This increased risk was also reflected by a significant association between cortisol secretion and facial fear perception, but only among subjects who experienced recurrence. A trend in the same direction was found for vocal fear perception. Fear perception and cortisol secretion per se did not predict recurrence. No gender differences were found. The association between cortisol secretion and fear perception (probably indicative for altered fear circuits in the brain) constitutes a mechanism underlying risk for recurrence of depression.