Temporal dynamics of reward anticipation in the human brain.

Reward anticipation is a complex process including cue evaluation, motor preparation, and feedback anticipation. The present study investigated whether these psychological processes were dissociable on neural dynamics in terms of incentive valence and approach motivation. We recorded EEG when participants were performing a monetary incentive delay task, and found a cue-P3 during the cue-evaluation stage, a contingent negative variation (CNV) during the motor-preparation stage, and a stimulus-preceding negativity (SPN) during the feedback-anticipation stage. Critically, both the cue-P3 and SPN exhibited an enhanced sensitivity to gain versus loss anticipation, which was not observed for the CNV. Moreover, both the cue-P3 and SPN, instead of the CNV, for gain anticipation selectively predicted the participants' approach motivation as measured in a following effort expenditure for rewards task, particularly when reward uncertainty was maximal. Together, these results indicate that reward anticipation consists of several sub-stages, each with distinct functional significance, thus providing implications for neuropsychiatric diseases characterized by dysfunction in anticipatory reward processing.

Zhang Y ，Li Q ，Wang Z ，Liu X ，Zheng Y ... -  《-》

An electrophysiological monetary incentive delay (e-MID) task: a way to decompose the different components of neural response to positive and negative monetary reinforcement.

The ability to anticipate and then secure future rewards and avoid future punishments by responding effectively to environmental demands is at the core of successful decision making. Disruptions to these processes have been shown to be implicated in a number of psychiatric conditions. In the current paper we use the electrophysiological monetary incentive delay task (e-MID) to decompose the neural response to (i) reinforcement anticipation, (ii) reinforcement-contingent target processing and (iii) reinforcement-related feedback. Thirty-eight adolescents and young adults performed an ERP-based analogue of the monetary incentive delay task. ERP components previously associated with motivationally salient cue (cue-P3 and contingent negative variation, CNV), target (P3) and feedback (success vs. failure; feedback-related negativity; FRN and the late positive potential; LPP) stimuli were examined. Response times were shorter and less variable in the monetary gain and loss conditions. Distinctive ERP components were observed for each phase of reinforcement processing. First, cue-P3 was enhanced to monetary gain cues. Predicted alterations in cue-P3 following monetary loss cues and the CNV following cues of either monetary loss or gain were not observed. Target P3 was enhanced in both incentive conditions. The FRN was greater following monetary loss feedback. LPP amplitude was enhanced following feedback denoting monetary gain and the avoidance of monetary loss. Although behaviourally the effects of monetary loss and gain were similar, the e-MID task differentiated neural processing in terms of anticipation and feedback-related brain potentials. The e-MID task and the results of the current study provide a valuable complement to fMRI-based approaches to studying normal and abnormal brain correlates of reinforcement processing.

Broyd SJ ，Richards HJ ，Helps SK ，Chronaki G ，Bamford S ，Sonuga-Barke EJ ... -  《-》

Teasing apart the anticipatory and consummatory processing of monetary incentives: An event-related potential study of reward dynamics.

The monetary incentive delay (MID) task has been widely used in fMRI studies to investigate the neural networks involved in anticipatory and consummatory reward processing. Previous efforts to adapt the MID task for use with ERPs, however, have had limited success. Here, we sought to further decompose reward dynamics using a comprehensive set of anticipatory (cue-N2, cue-P3, contingent negative variation [CNV]) and consummatory ERPs (feedback negativity [FN], feedback P3 [fb-P3]). ERP data was recorded during adapted versions of the MID task across two experiments. Unlike previous studies, monetary incentive cues modulated the cue-N2, cue-P3, and CNV; however, cue-related ERPs and the CNV were uncorrelated with one another, indicating distinct anticipatory subprocesses. With regard to consummatory processing, FN amplitude primarily tracked outcome valence (reward vs. nonreward), whereas fb-P3 amplitude primarily tracked outcome salience (uncertain vs. certain). Independent modulation of the cue-P3 and fb-P3 was observed, indicating that these two P3 responses may uniquely capture the allocation of attention during anticipatory and consummatory reward processing, respectively. Overall, across two samples, consistent evidence of both anticipatory and consummatory ERP activity was observed on an adapted version of the MID paradigm, demonstrating for the first time how these ERP components may be integrated with one another to more fully characterize the time course of reward processing. This ERP-MID paradigm is well suited to parsing reward dynamics, and can be applied to both healthy and clinical populations.

Novak KD ，Foti D 《-》

Motivated to win: Relationship between anticipatory and outcome reward-related neural activity.

Reward-processing involves two temporal stages characterized by two distinct neural processes: reward-anticipation and reward-outcome. Intriguingly, very little research has examined the relationship between neural processes involved in reward-anticipation and reward-outcome. To investigate this, one needs to consider the heterogeneity of reward-processing within each stage. To identify different stages of reward processing, we adapted a reward time-estimation task. While EEG data were recorded, participants were instructed to button-press 3.5s after the onset of an Anticipation-Cue and received monetary reward for good time-estimation on the Reward trials, but not on No-Reward trials. We first separated reward-anticipation into event related potentials (ERPs) occurring at three sub-stages: reward/no-reward cue-evaluation, motor-preparation and feedback-anticipation. During reward/no-reward cue-evaluation, the Reward-Anticipation Cue led to a smaller N2 and larger P3. During motor-preparation, we report, for the first time, that the Reward-Anticipation Cue enhanced the Readiness Potential (RP), starting approximately 1s before movement. At the subsequent feedback-anticipation stage, the Reward-Anticipation Cue elevated the Stimulus-Preceding Negativity (SPN). We also separated reward-outcome ERPs into different components occurring at different time-windows: the Feedback-Related Negativity (FRN), Feedback-P3 (FB-P3) and Late-Positive Potentials (LPP). Lastly, we examined the relationship between reward-anticipation and reward-outcome ERPs. We report that individual-differences in specific reward-anticipation ERPs uniquely predicted specific reward-outcome ERPs. In particular, the reward-anticipation Early-RP (1-.8s before movement) predicted early reward-outcome ERPs (FRN and FB-P3), whereas, the reward-anticipation SPN most strongly predicted a later reward-outcome ERP (LPP). Results have important implications for understanding the nature of the relationship between reward-anticipation and reward-outcome neural-processes.

Pornpattananangkul N ，Nusslock R 《-》

Electrocortical components of anticipation and consumption in a monetary incentive delay task.

In order to improve our understanding of the components that reflect functionally important processes during reward anticipation and consumption, we used principle components analyses (PCA) to separate and quantify averaged ERP data obtained from each stage of a modified monetary incentive delay (MID) task. Although a small number of recent ERP studies have reported that reward and loss cues potentiate ERPs during anticipation, action preparation, and consummatory stages of reward processing, these findings are inconsistent due to temporal and spatial overlap between the relevant electrophysiological components. Our results show three components following cue presentation are sensitive to incentive cues (N1, P3a, P3b). In contrast to previous research, reward-related enhancement occurred only in the P3b, with earlier components more sensitive to break-even and loss cues. During feedback anticipation, we observed a lateralized centroparietal negativity that was sensitive to response hand but not cue type. We also show that use of PCA on ERPs reflecting reward consumption successfully separates the reward positivity from the independently modulated feedback-P3. Last, we observe for the first time a new reward consumption component: a late negativity distributed over the left frontal pole. This component appears to be sensitive to response hand, especially in the context of monetary gain. These results illustrate that the time course and sensitivities of electrophysiological activity that follows incentive cues do not follow a simple heuristic in which reward incentive cues produce enhanced activity at all stages and substages.

Angus DJ ，Latham AJ ，Harmon-Jones E ，Deliano M ，Balleine B ，Braddon-Mitchell D ... -  《-》