Outcome Evaluation in Decision Making: ERP Studies

Abstract

Decision making can be regarded as an outcome of mental processes (cognitive process) leading to the selection of a course of action among several alternatives. Every decision-making process produces a final choice. In this chapter, we would use the event-related brain potential (ERP) technique to study neural correlates of outcome evaluation in decision making. Firstly, we used the motivation to be deceptive as a way to manipulate the motivational state of participants. Results showed that in both groups, losing money evoked a negative component (FRN). The amplitude of FRN was higher, and the latency shorter, in the deception group than in the simple response group. In addition, monetary losses also elicited a P3 peaking at around 400 ms, which was larger in the deception group than in the simple response group. Source modeling suggested a dorsal anterior cingulate cortex (ACC) source for the FRN and a rostral ACC source for the P300. These results suggest that motivation can influence the evaluation of performance outcomes. The FRN may reflect a process signaling that the outcome is worse than what participants expect, and P300 may be related to the emotional processing of disappointment and regret. Secondly, in a simulated deception situation with graded monetary incentives, participants made a decision to lie or be truthful in each trial and held their response until a delayed imperative signal was presented. Spatiotemporal principal component analysis (PCA) and source analysis revealed that brain activities dominant in the left lateral frontal area approximately 800-1000 ms post-stimulus and over the central-frontal-parietal and right frontal areas after 1300 ms were significantly more negative in the deceptive condition than in the truthful condition. These results suggest that two serial cognitive processes, decision-making and response preparation, are related to deliberate deception. Thirdly, an ERP experiment was conducted to investigate the component processes underlying error related feedback and regret following incorrect decisions. Results showed that the amplitude of the FRN following an incorrect choice, the personal responsibility people accepted for an incorrect choice, the regret they expressed about it, and the counterfactual thinking they reported were higher when they alone made the choice than when they were not alone in their incorrect choice. No differences were found on the FRN or behavioral measures when participants were joined by one versus two others in their incorrect choice. The P300 amplitude, in contrast, was inversely related to the responsibility levels, consistent with the notion that personal accountability of a choice influences the allocation of attentional and cognitive resources to the task. Together, these results are consistent with neural modeling indicating that regret promotes error related processing and learning. Lastly, Compared with non-anxious people, anxious people are suggested to judge the negative events as more likely to happen, and more likely to interpret the ambiguous outcomes as negative ones. We predicted that the FRN should be different between high trait-anxiety (HTA) and low trait-anxiety (LTA) groups. We discovered that the amplitude of the FRN indicating negative versus positive outcomes was significantly larger for LTA participants than HTA participants. However, the intergroup difference of the FRN indicating ambiguous versus positive outcomes didn't reach significance. The result indicated that there was a relationship between the FRN and the participants' individual differences in anxiety, which possibly reflected the impact of anxiety on outcome expectation. This finding provided insight about the underlying features of anxiety.