责任归因对“做效应”的调控及其ERP证据

Modulation of Attribution of Responsibility on ‘Action Effect’ and Its ERP Evidence

"做效应(actioneffect)"是指决策中由"做(action)"与"不做(inaction)"行为引起的相同决策后果,却诱发了不同强度的情绪体验。本研究使用ANSIE量表筛选了责任归因不同倾向的两类被试(极端内控者、极端外控者),运用ERP技术,在一项简单赌博实验任务中考察了这两类被试的"做效应"差异及其电生理证据,旨在探讨责任归因对"做效应"的调控作用。行为结果表明,无论结果输赢,极端内控被试的"做"与"不做"行为结果诱发的情绪强度和责任感强度都没有差异,而极端外控被试的"做"行为诱发的情绪强度和责任感强度都比"不做"行为诱发的明显大;脑电结果表明,在FRN和P300上,极端内控被试的"做"与"不做"行为诱发的FRN和P300波幅之间都没有差异,而极端外控被试的"做"行为诱发的这两种脑电成分波幅都比"不做"诱发的较大。简言之,"做效应"在极端内控被试身上表现不明显,而在极端外控被试身上表现明显。"做效应"很可能是由于个体对不同行为导致的相同决策后果的责任归因不同引起的。

The relationship between emotion and decision-making is of significant interest in the field of scientific psychology as well as in economics research. The emotions that individuals experience for post-decisional outcomes are not only influenced by properties of the outcomes （e.g. magnitude, valence）, but also by the paths of individuals＇ action （e.g., action vs. inaction） to reach it. Considerable behavioral researches have claimed that emotional reactions to outcomes following decisions to act are typically more intense than are those following decisions not to act. This well-known phenomenon has been labeled ＇action effect＇ （Landman, 1987, Kahneman and Tversky, 1982）. Though numerous convergent studies on decision-making have reported this phenomenon, few have reached agreement on its cognitive mechanism. By simultaneously recording event-related potentials （ERPs） data from participants with different locus of control （internal versus external locus of control, ILC vs ELC ） engaged in a simple gambling task, the present study aimed to further investigate the role of attribution of responsibility in inducing the ＂action effect＂. Fifteen ILC participants （7 males and 8 females, age18-25, M=20.69_＋2.28 years） and fourteen ELC participants （5 males and 9 females, age17-25, M=20.42_＋2.19years） took part in the experiment. The two groups were selected from 256 volunteers recruited from Southwest University in China with the Adult Nowicki-Strickland Internal-External Control Scale （ANSIECS）. The participants in the experiment were all right-hand, had normal or corrected-to-normal vision and had no neurological or psychological disorders. Each participant signed a consent form prior to the experiments and was paid after the experiment, according their performance. The study was approved by the local academic committee. In this experimental task, participants were first presented with two choice options and were told that the computer would choose one of two options for them. Both choices were randomly associated with identical monetary gains and losses in each trial. Participants were asked to decide whether to keep the computer choice （inaction condition） or to choose the other one （action condition）. Subsequently, a feedback conveying the outcomes of the two options （either gaining or losing money） was provided. By manipulating the action types that depended on whether the participants followed the chosen options by the computer and the valence of the feedback in the gambling task, four conditions were included： inaction-gain; inaction-loss; action-gain; and action-loss. The behavioral data showed that, whatever the feedback outcomes were winning or losing, the emotions and responsibility experienced by the ILC participants for the feedback outcomes were not significantly different between following their action and inaction; while the emotions and responsibility experienced by the ELC participants for the feedback outcomes following their action were more intense than that following their inaction. Moreover, the FRNs and P300s elicited by the feedback outcomes following action and inaction were no differences for the ILC participants, whereas the two ERP components elicited by the feedback outcomes following action were both larger than that following inaction for the ELC participants. In short, the action effect is not evident in ILC participants, but evident in EIC participants. Thus, it most likely that action effect is induced by different degree of sense of responsibility to identical outcomes following variant behavior.