Effects Of Reward Valence On Anticipatory Phase Of Reward Processing

Aim:Reward processing is crucial for human survival.Alterations in the reward system have been associated with various neuropsychiatric disorders.Previous studies have treat reward system as a single structure.However,reward can be decomposed into two main successive phases:the anticipation phase and the outcome-appraisal phase.The event-related potential(ERP)technique,with its superior temporal resolution,provides a unique opportunity to examine the neural dynamics of reward processing,which permits the separation of neural processes that occur very closely in time.The ERP component reflecting outcome-appraisal processing has attracted lots of attention and has been used as a neurobiology index in research on neuropsychiatric disorders.Recent evidence points to the hypothesis that the stimulus-preceding negativity(SPN)may constitute an electrophysiological index of reward anticipation.However,it remains controversial whether the anticipatory process indexed by the SPN is modulated by reward valence.The present study investigated the effect of valence on the SPN in a gambling task that required participants to make a binary(gain vs.loss)prediction after their choice on trials with different reward probabilities.Moreover,the inclusion of reward probabilities in the current experiment allowed us to examine the potential effect of objective valence on the SPN.Methods:39 normal participants were recruited in this study.The participants performed a simple gambling task that asked the participants to select one of the four doors on trials with different reward probabilities.Following their response,the participants were asked to make a prediction whether the chosen door would be rewarded or punished,thereafter,a feedback stimulus appeared to inform them whether the outcome was win or lose on that trial.The formal experiment task consisted of 400trials divided into five blocks.Before the formal experiment,a practice version of the task was utilized to familiarize the participants with the task.The EEG data were recorded using Neuroscan SynAmp~2 and preprocessed using MATLAB and EEGLAB toolbox.In addition,all the participants were asked to complete a questionnaire about the task.All statistical analyses were performed using IBM SPSS(v19.0).Firstly,we compared the participants’estimation score about the task.Secondly,we analyzed the reaction time and the gain prediction and estimation across the three reward probability levels.Finally,we performed two analyses for the subjective and objective SPN data,respectively.Results:1.The score of participants’interest was significantly higher than the standard(t=9.52,p<0.001),and the participants drew more attention during the prediction(t=5.23,p<0.001)and the outome(t=8.58,p<0.001).Moreover,the pleasure score of gain was significantly higher than that of lose(t=10.53,p<0.001).2.Compared to the objective gain probability,the percentage of gain prediction was more than 25%on 1-cue trials,t(38)=3.06,p<.01,more than 50%on 2-cue trials,t(38)=5.13,p<.001,and more than 75%on 3-cue trials,t(38)=15.87,p<.001,indicating a generally positive bias tendency across all the three cue types.Moreover,this positive bias was modulated by objective gain probability.Post hoc comparisons indicated that the positive bias was enhanced on 2-cue and 3-cue trials compared to1-cue trials(ps<.05),with no difference between 2-cue and 3-cue trials(p>.05).3.A significant three-way interaction among subjective valence,hemisphere,and site was observed,F(2,76)=4.27,p=.024,ηp2=.10.Post hoc comparisons indicated that the SPN was greater in response to predicted gain compared to predicted loss only at T7(-1.46 vs.0.07μV,p=.018).4.The interaction of objective valence and uncertainty was marginally significant,F(1,38)=3.53,p=.068,ηp2=.09.Post hoc comparisons indicated that the uncertainty effect was more pronounced in objective gain anticipation(ΔM=-2.32μV,p<.001).Conclusions:1.The enhanced sensitivity to positive versus negative valence was found during the anticipatory phase of reward processing.2.The SPN was enhanced before uncertain versus certain outcomes,which was more pronounced during objective gain relative to loss anticipation.3.There is a valence asymmetry in reward anticipation as indexed by the SPN,which may be mediated by the mesocorticolimbic dopaminergic pathway.