The Effect Of Construal Level On Outcome Evaluation:An ERP Study

People always use a reference point that is the self, here and now when processing object event. However, when the object happens to someone else, in another place, at a distant point in time, or when imaginary, what will happen to the representations of the object? According to construal level theory (CLT), psychological distance can affect people’s psychological construction or representation of the object event. Psychologically distant events are represented in terms of abstract, general, superordinate, and decontextualized features that emphasize essentialities (high-level construals), whereas psychologically proximal events are represented in terms of concrete, incidental, subordinate, and contextual features that incorporate peripheral concerns (low-level construals). Numberous behavioral evidence showed that manipulations of construal levels can impact people’s inter-temporal choice and risky choice. In general, as compared to low-level construals, high-level construal individuals prefer to the long-term goal or risk option.Why manipulations of construal levels can influence people’s risk-taking behavior? Few studies speculated that different levels of construals can change individuals’cognitive or emotional experience of decision problems. In our opinion, manipulations of construal levels not only influence selection stage of decision making, but also influence outcome evaluation by changing cognitive or emotional experience following decision outcome. As compared to high-level construals, the outcome of low-level construal events elicits larger prediction errors and more intense emotional experience. In order to test this hypothesis, we used ERPs to explore the effect of construal level on the time course of outcome evaluation. We focused on two ERP components typically associated with outcome evaluation, the feedback-related negativity (FRN) and the P300. According to relevant literature, our research logic is: the FRN represents the transmission of a prediction error signal and reflect motivational impact of the current event. P300 is associated with top-down emotional motivation process. If manipulations of construal level and psychological distance change the cognitive or emotional experience following decision outcome, this process can also influence the FRN and P300 amplitudes elicited by outcome feedback.This thesis included three studies:study 1 used two experiments to explore the effect of construal level on the electrophysiological correlates of outcome feedback. Previous literatures have demonstrated that priming attentional scope and mindset is two kinds of effective way to manipulate construal level. Therefore, in the experiment 1, we used navon letters (1977) to priming scope of attention, global letters elicited high-level construals and local letters elicited low-level construals, and then investigate the time course of outcome feedback in a gambling task. Experiment 2 required participants to consider questions related to "why" and "how"they engaged in certain actions which priming high or low level construal mindsets, and then investigate the time course of outcome feedback in a gambling task. Owing to the relationship between construal level and psychological distance is bidirectional, study 2 used three experiments to explore the effect of psychological distance on the electrophysiological correlates of outcome feedback. Specifically, experiment 3 required participants to gambling for oneself, friends, or strangers in a gambling game, investigate how social distance influence outcome feedback. Experiment 4 required participants to gambling for now or for 1 month later in a gambling game, investigate how temproal distance influence outcome feedback. In experiment 5, the gambling stimuli were presented within the Ponzo lines, in a "distal" (upper) or "proximal" (lower) locations, investigate how spatial distance influence outcome feedback. In order to apply CLT to more complex field of outcome evaluation area, Study 3 used dot estimation task and required participants to compare with average or concrete others’feedback respectively, then presented three social comparison direction (upward comparison/disadvantageous inequity, parallel comparison/equity, downward comparison/advantageous inequity) as reward type, investigate how construal level influence reward processing in the social comparison context.The results of study 1 showed that both priming attentional scope and mindset to manipulation construal level can influence the time course of outcome feedback. Experiment 1 found that as compared to a global attentional scope, a local attentional scope increased both the dFRN and the P300 amplitudes.Experiment 2 found that as compared to consider "why" questions, consider "how" questions increased the dFRN amplitudes. The interaction effect between mindset and feedback valence was significance:gain condition elicted larger P300 amplitudes than loss condition after consider "how" questions. But there was no difference in P300 amplitudes between loss and gain conditions after consider "why"questions.The results of study 2 showed that manipulations of psychological distance can influence the time course of outcome feedback. Experiment 3 found that as compared to gambling for strangers, gambling for oneself increased both the dFRN and the P300 amplitudes. Experiment 4 found that as compared to gambling for 1 month later, gambling for now increased both the dFRN and the P300 amplitudes. Experiment 5 found that as compared to distal feedback stimuli, proximal feedback stimuli elicited larger dFRN amplitudes, but there was no difference in P300 amplitudes.The results of study 3 showed that construal level can influence the electrophysio logical correlates of reward processing in the social comparison context. Experiment 6 found that in the concrete others condition, disadvantageous inequity condition elicited larger FRN amplitudes as compared to advantageous inequity and equity conditions. The FRN amplitudes between advantageous inequity and equity conditions were not significant. In the average others condition, The FRN amplitudes among three conditions were not significant. Construal level was not influence P300 amplitudes elicited by reward type.In sum, manipulations of construal level and psychological distance both influence the FRN reflected in reinforcement learning process, suggesting these manipulations modulate the earlier semi-automatic outcome evaluation process, low-level construals and proximity psycho logical distance elicit larger prediction error and more intensity motivation to win money. Whereas P300 component may be due to the different methods of manipulation, the conclusion is slightly different. The results not only enrich the research contents in the field of reinforcement learning, but also provide a new research idea to consider how construct level influence decision-making.