Objective Major depressive disorder (MDD) is characterized by impaired reward processing, possibly due to dysfunction in the basal ganglia. However, few neuroimaging studies of depression have distinguished between anticipatory and consummatory phases of reward processing. Using functional magnetic resonance imaging (fMRI) and a task that dissociates anticipatory and consummatory phases of reward processing, the authors tested the hypothesis that MDD participants would show reduced reward-related responses in basal ganglia structures. Method A monetary incentive delay task was presented to 30 unmedicated MDD subjects and 31 healthy comparison subjects during fMRI scanning. Whole-brain analyses focused on neural responses to reward-predicting cues and rewarding outcomes (i.e., monetary gains). Secondary analyses focused on the relationship between anhedonic symptoms and basal ganglia volumes. Results Relative to comparison subjects, MDD participants showed significantly weaker responses to gains in the left nucleus accumbens and bilateral caudate. Group differences in these regions were specific to rewarding outcomes and did not generalize to neutral or negative outcomes, although relatively reduced responses to monetary penalties in MDD emerged in other caudate regions. By contrast, evidence for group differences during reward anticipation was weaker, although MDD subjects showed reduced activation to reward cues in a small sector of the left posterior putamen. Among MDD subjects, anhedonic symptoms and depression severity were associated with reduced bilateral caudate volume. Conclusions These results indicate that basal ganglia dysfunction in MDD may affect the consummatory phase of reward processing. Additionally, morphometric results suggest that anhedonia in MDD is related to caudate volume.

The authors examined extreme response style in recurrently and chronically depressed patients, assessing its role in therapeutic outcome. During the acute phase, outpatients with major depressive disorder (N ϭ 384) were treated with fluoxetine for 8 weeks. Remitted patients (n ϭ 132) entered a continuation phase during which their fluoxetine dose increased and they were randomly assigned to treatment with or without cognitive-behavioral therapy (CBT). Results showed a predictive relationship between extreme response style and clinical outcome. Patients in the medication-only group showed a significant increase in the frequency of extreme responses, whereas patients receiving CBT showed no significant change. These results are consistent with recent findings suggesting that metacognitive factors may be as important as changes in thought content when treating depression.

No abstract