Background: We assessed the therapeutic effects of bupropion SR and placebo on mood and anxiety symptoms derived from the tripartite model of mood. Based on evidence indicating linkages between dopaminergic activity and the emotional dimension of positive affect/anhedonia, we hypothesized that the dopaminergic effects of bupropion SR would yield particularly pronounced effects on symptoms of anhedonia, relative to anxiety. Methods: Nineteen depressed outpatients were randomly assigned to treatment with either bupropion SR 300 mg/day or placebo during a 6-week initial treatment phase. This was followed by a second open-label phase in which patients previously treated with bupropion SR had their dose increased to 400 mg/day, and the placebo group was initiated on bupropion SR 300 mg/day. Results: Random regression analyses revealed that during the initial double-blind phase, bupropion SR elicited greater declines than placebo on all measures except those that assessed anxiety. By contrast, the weakest placebo effects were evident on anhedonia. Items assessing the low positive affect pole of the anhedonia dimension were more sensitive to earlier/lower dose bupropion SR treatment, whereas items assessing the high positive affect pole were more sensitive to later/higher dose bupropion SR treatment. Limitations: Replication and extension using a larger sample size are mandated. Conclusions: This study suggests that the catecholaminergic effects of bupropion SR tended to produce more robust effects on anhedonia/positive affect than placebo. © 2004 Elsevier B.V. All rights reserved.