OBJECTIVE: The present study aimed to 1) assess facial expression recognition in subjects with a previous history of major depressive disorder relative to subjects with no history of depression and 2) characterize the effects of acute citalopram infusion on recognition performance for both groups. METHOD: Unmedicated euthymic women with a history of major depression and matched comparison subjects with no history of depression were given a facial expression recognition task following intravenous infusion of saline or citalopram (10 mg) in a double-blind, between-group design. RESULTS: Following saline infusion, subjects with a previous history of depression showed a selectively greater recognition of fear relative to the subjects with no history of depression. The abnormal fear processing observed in the subjects with a previous history of depression was normalized following citalopram infusion, an effect that was opposite to that seen with the subjects with no history of depression. CONCLUSIONS: These results suggest that increased recognition of fear is a trait vulnerability marker for depression and that this is normalized following a single dose of citalopram.