Blunted Neural Effects Of Perceived Control On Reward Feedback In Major Depression:An Event-Related Potential Study

Objective:A wealth of research has linked depression to deficits in reward sensitivity.Blunted reward processing has emerged as an endophenotype of major depressive disorder(MDD).However,mechanistic understanding for this deficit remains ambiguous.Combining the recent evidence emphasizes that being able to exert control over the environment can boost the influences of positive information by increasing the activation of the reward network in the brain.We speculate that the mechanism of reward processing deficits in major depressive disorder is driven by the blunted neural effect of perceived control on reward processing.Event-related potentials(ERPs)can be used to accurate to the millisecond and to reflect the dynamic process of cognitive processing.This study intends to use ERPs technology to preliminarily investigate the relationship between the blunted neural effect of perceived control and reward processing deficits in MDD patients.Methods:This study adapted a well-validated gambling task(the doors task)in which perceived control is exercised by choice in 29 individuals with current MDD and31 healthy controls while their EEG was recorded.After off-line analysis,the reward positivity was quantified across frontocentral electrodes(i.e.,Fz,FC1,FCz,FC2)as the mean activity.The P300 was measured as the mean activity across centroparietal electrodes(i.e.,C1,Cz,C2,CPz).Rew P and P300 amplitudes were calculated under four experimental conditions(high control-gain、high control-nongain、low control-gain、low control-nongain).A mixed repeated-measures analysis of variance(ANOVA)was used separately for reaction time(RT)data,post-experimental rating data,and ERP data.Results:1.Hamilton depression scale score(23.41±3.69)in MDD patients was higher than that in healthy control group(2.81±1.60)(p<0.001).The Hamilton anxiety scale score(19.90±6.63)was higher than that of the healthy control group(2.07±1.53)(p<0.001).2.The reward positivity was larger in the choice condition(M=1.52±0.23μV)than in the no-choice condition(M=0.95±0.20μV),F(1,58)=6.17,p=0.016,ηp2=0.10,and for healthy controls(M=1.79±0.26μV)than for individuals with MDD(M=0.68±0.27μV),F(1,58)=8.88,p=0.004,η_p2=0.13.Importantly,the main effects were qualified by a significant interaction between group and choice,F(1,58)=4.60,p=0.036,η_p2=0.07.Post hoc pair wise comparisons showed that reward positivity amplitude was significantly increased in the choice condition compared to the no-choice condition in healthy controls(p=0.002)but not in MDD participants(p=0.814).Further,the reward positivity was larger for healthy controls than for individuals with MDD in the choice condition(p=0.001)but not in the no-choice condition(p=0.131).Thus,the clearfinding here was that healthy participants were able to experience the choice effect on the reward positivity,whereas participants with MDD were not.3.The P300 was larger in the choice condition(M=4.83±0.31μV)than in the no-choice condition(M=2.88±0.22μV),(F(1,58)=54.95,p<0.001,η_p2=0.49),and following gain feedback(M=4.52±0.24μV)than following nongain feedback(M=3.19±0.26μV),(F(1,58)=62.18,p<0.001,η_p2=0.52).Healthy controls(M=4.48±0.33μV)exhibited a larger P300 than participants with MDD(M=3.24±0.34μV),(F(1,58)=6.93,p=0.011,η_p2=0.11).However,this group effect was qualified by a significant interaction between group and valence(F(1,58)=6.70,p=0.010,η_p2=0.11).Post hoc analyses revealed that healthy participants elicited a greater P300 than MDD participants when feedback was gain(p=0.001),but not when it was nongain(p=0.136).No other interaction effects were significant(p>0.2).Conclusions:1.A high perceived control promotes reward processing in healthy controls.2.MDD patients have deficits in reward processing occurs only under conditions of high perceived control.3.Reward processing deficits in MDD patients may be due to their abnormal perceived control.