Abnormal reward system activation in mania

Abstract

Transmission of reward signals is a function of dopamine, a neurotransmitter known to be involved in the mechanism of psychosis. Using functional magnetic resonance imaging (fMRI), we investigated how expectation and receipt of monetary rewards modulate brain activation in patients with bipolar mania and schizophrenia. We studied 12 acutely manic patients with a history of bipolar disorder, 12 patients with a current episode of schizoaffective disorder or schizophrenia and 12 healthy subjects. All patients were treated with dopamine antagonists at the time of the study. Subjects performed a delayed incentive paradigm with monetary reward in the scanner that allowed for investigating effects of expectation, receipt, and omission of rewards. Patients with schizophrenia and healthy control subjects showed the expected activation of dopaminergic brain areas, that is, ventral tegmentum activation upon expectation of monetary rewards and nucleus accumbens activation during receipt vs omission of rewards. In manic patients, however, we did not find a similar pattern of brain activation and the differential signal in the nucleus accumbens upon receipt vs omission of rewards was significantly lower compared to the healthy control subjects. Our findings provide evidence for abnormal function of the dopamine system during receipt or omission of expected rewards in bipolar disorder. These deficits in prediction error processing in acute mania may help to explain symptoms of disinhibition and abnormal goal pursuit regulation.

Figure 1

Mean reaction times and standard deviation during scanning of the reward task in the three groups (for the three trial types: high/low/no reward). In the high reward trials of the controls and SZ patient groups, we observed a significant (*p<0.05) acceleration of reaction times compared to both the low and no reward trials, which was not present (○) in the manic patients.

Figure 2

Differential fMRI activation (controls (outcome: win>omission of reward) >mania (outcome: win>omission of reward)) at p<0.005, uncorrected for multiple comparisons.

Figure 3

Functional magnetic resonance imaging (fMRI) activation (mania, schizophrenia, controls groups together, contrast: expectation of high4expectation of no reward) and time courses (first eigenvariate of the fMRI signal intensity as provided by standard SPM functions) in left and right ventral tegmental area for each of the groups. The map was thresholded at p<0.001 and extent threshold corrected at p<0.05. Mean-corrected first eigenvariate values were extracted from the left and right ventral tegmental functional ROI for each subject and averaged over groups. The time course data were averaged event-related to depict the mean fMRI signal and standard error related to expectation of high and no reward in each of the three groups. Grey shades indicate the period when reward expectation and receipt or omission of reward took place relative to the MR signal, assuming a delay of the BOLD signal of 6 s. ⇑ : Significant differences (p<0.05) between conditions. Mania: bipolar manic patients; Schizo.: schizophrenia/schizoaffective patients; Cont.: healthy control subjects.

Figure 4

(a) Differential functional magnetic resonance imaging (fMRI) activation (Mania, Schizophrenia, Controls groups together, contrast: outcome: win>omission of reward) and time courses (first eigenvariate of the fMRI signal intensity as provided by standard SPM functions) in left and right nucleus accumbens (NAcc) for each of the groups. The map was thresholded at p<0.001 and extent threshold corrected at p<0.05. Mean-corrected first eigenvariate values were extracted from the left and right NAcc functional ROI for each subject and averaged over groups. The time course data were averaged event-related to depict the mean fMRI signal and standard error related to the outcome phase (win and omission of reward), in each of the three groups. Grey shades indicate the period when reward expectation and receipt or omission of reward took place relative to the MR signal, assuming a delay of the BOLD signal of 6 s. ⇑: significant differences (p<0.05) between conditions. ⇓: significantly (p<0.05) greater difference (win>omission of reward)in controls than in mania. (b) Mean values of modeled effects in each of the three groups for the contrast (outcome: win>omission of reward) from the regions of interest (ROI) defined from the NAcc activations of the overall analysis of the three groups together. Mania: bipolar manic patients; Schizo.:schizophrenia/schizoaffective patients: Cont.: healthy control subjects.