The effects of psychotherapy on neural responses to rewards in major depression

Abstract

Background: Unipolar major depressive disorder (MDD) is characterized by anomalous neurobiological responses to pleasant stimuli, a pattern that may be linked to symptoms of anhedonia. However, the potential for psychotherapy to normalize neurobiological responses to pleasant stimuli has not been evaluated.

Methods: Twelve adults with and 15 adults without MDD participated in two identical functional magnetic resonance imaging scans that used a Wheel of Fortune task. Between scans, MDD outpatients received Behavioral Activation Therapy for Depression, a psychotherapy modality designed to increase engagement with rewarding stimuli and reduce avoidance behaviors.

Results: Seventy-five percent of adults with MDD were treatment responders, achieving post-treatment Hamilton Rating Scale for Depression score of six or below. Relative to changes in brain function in the matched nondepressed group, psychotherapy resulted in functional changes in structures that mediate responses to rewards, including the paracingulate gyrus during reward selection, the right caudate nucleus (i.e., the dorsal striatum), during reward anticipation, and the paracingulate and orbital frontal gyri during reward feedback. There was no effect of diagnostic status or psychotherapy on in-scanner task-related behavioral responses.

Conclusions: Behavioral Activation Therapy for Depression, a psychotherapy modality designed to increase engagement with rewarding stimuli and reduce avoidance behaviors, results in improved functioning of unique reward structures during different temporal phases of responses to pleasurable stimuli, including the dorsal striatum during reward anticipation.

Figures

Figure 1

The Wheel of Fortune task; (a) The four different wheel types; (b) The timing of three task processes, i.e., reward selection, reward anticipation, and reward feedback.

Figure 2

Top Left: Percentage of high risk selections (i.e., smaller pie slices) for the 10/90 and 30/70 conditions for both diagnostic groups at both timepoints. Top Right. Average confidence ratings (range = 1 to 6) for both diagnostic groups at both timepoints. Note that the unequal wheels are subdivided based on selection selections on a trial-by-trial basis (i.e., “risky” or “safe” selections). Bottom Left: Ratings of feedback valence for win trials for both diagnostic groups at both timepoints. On win trails, the range and direction of the ratings were 1 = Neutral, 6 = Very Happy, whereas on no-win trials, the range and direction of the ratings were 1 = Very Sad, 6 = Neutral. Bottom Right: Ratings of feedback valence for non-win trials for both diagnostic groups at both timepoints. On win trails, the range and direction of the ratings were 1 = Neutral, 6 = Very Happy, whereas on no-win trials, the range and direction of the ratings were 1 = Very Sad, 6 = Neutral.

Figure 3

Reward selection (top) and anticipation (bottom) fMRI results (money vs. control). Activation images denote voxels with significant Group (Depressed, Nondepressed) X Time (Time 1, Time 2) interactions. Neurological convention (right on right) is used and coordinates are in MNI space. Signal intensity is T statistic units.

Figure 4

Reward feedback fMRI results for win (top) and non-win (bottom) trials, relative to control feedback. Activation images denote voxels with significant Group (Depressed, Nondepressed) X Time (Time 1, Time 2) interactions. Neurological convention (right on right) is used and coordinates are in MNI space. Signal intensity is T statistic units.