Research On Cognitive Mechanisms Of Chosen Lie And Instructed Lie

The lying behavior is not only a widespread social phenomenon, but also a complex cognitive process. As early as the beginning of the20th century, Jean Piaget studied the development of children’s lying and moral judgment, but so far we have been unclear about the psychological process of lying.For the past10years, researchers have made progress in the exploration of lying, using EPR, fMRI and other noninvasive brain imaging means. However, they cared little about the cognitive processes and the neural mechanism of lying, especially through ERP. Resent ERP studies on lying have focused on two components:P300wave reflecting stimulus classification and MFN reflecting conflict monitoring. Nevertheless, little EPR research is related to the processing components of cognitive conflicts and control involved in lying, and we are still unclear about the cognitive processes of lying. Therefore, for the purpose of exploring and understanding the complex cognitive process and corresponding neural mechanisms of lying behavior more clearly, we divided it into two parts-the motive stage and the action stage by behavioral experiments, combined with the ERP technology.Meanwhile, studies on lying mostly used the experimental paradigm that participants make lying reaction under the requirements of the experimenters, rather than make their choice of lying or not lying. This experimental paradigm is different from the lying behavior in the real life, probably reducing the ecological validity of the experiment. Therefore, in order to explore the possible differences between chosen lie and instructed lie, we intended to adopt a paradigm similar to Carrion face-to-face liar game, and we changed the human-human interaction into the human-computer one. This could exclude possible moral conflicts, and thereby discuss the pure cognitive processing mechanisms.Our study included three experiments of two behavioral experiments and an ERP experiment.In Study1, we adopted a paradigm similar to Carrion face-to-face liar game, containing two components, which were reaction types of choosing and instructed behaviors, and reaction properties of lying and truth-telling behaviors, thus constituting four conditions of instructed lie, instructed truth, chosen lie and chosen truth. Results showed that interaction effect between reaction types and properties was significant. That is, in the chosen reaction condition, there was no marked difference between RTs of lie and truth; in the instructed reaction condition, RT of lying was less than that of truth-telling. This discordance of these two reaction types demonstrated that the cognitive mechanism of choosing reaction was probably different from that of instructed reaction, which was consistent with our hypothesis. But there existed some problems experiment about the experiment. No reaction option under the choosing reaction condition reduced the validity of this experiment. Therefore, for the purpose of intensive study, we conducted Study2.In Study2, we improved our experimental paradigm. Participants were required to react twice in one trial. The first stage was the motive stage in which participants confirmed whether they intended to lie. The second stage was the action stage in which participants implemented responses that they just decided to make. Only the trails of which both reactions were right were entered the statistical analysis, which could avoid participants trend of pressing keys randomly under the choosing condition in order to save time, and study the cognitive process of lying more deeply by stages. Results showed that during the motive stage, there was significant interaction effect between reaction types and properties. That is, there was no difference between RTs of chosen lie and chosen truth, while RT of instructed lie was less than that of instructed truth. This indicated that we might produce some cognitive conflicts when we only realized that we would lie next. During the action stage, the main effect of reaction properties that RTs under the lying condition were saliently less than those under the truth-telling condition, was significant. Thus we can speculate that during the reaction stage, the implementation of responses may also have some kind of conflict, and this conflict is obviously different from that during the motive stage. Additionally, the interaction effect between the reaction properties and types was significant, and the trends of these two reaction types were inconsistent with each other, demonstrating reaction types did not only affect the motive stage. According to these finding, we needed to seek further proof through the ERP technology.The process of Study3was basically the same as that of Study2, but the interval time was adjusted to the ERP experiments. The RT data of Study3were also basically consistent with those of Study2. The ERP data indicated that during the motive stage, only under the instructed condition could there be obvious N2and P3components. We found that the N2amplitude under instructed lie condition was significantly less than that under instructed truth condition. As the frontal N2wave is related to cognitive conflicts, this finding explains that the difference between RTs of instructed lie and truth was indeed the result of cognitive conflict in Study1. During the action stage, the main effect of reaction properties through P3and N4components were significant, which was consistent with previous research on lying, but was obviously inconsistent with RT differences resulted from N2amplitude differences during the motive stage. This showed that there should be differences in at least three cognitive components between lie and truth. Meanwhile, during the action stage, the main effects of reaction types in P3and N4components were significant, indicating that these two reaction types had the same cognitive processing mechanism, but were different in the cognitive level. Besides, there was another difference under the chosen condition. That is, the N2amplitude in chosen lie condition was marginally different from that in chosen truth condition, but the gap of RT between these two conditions was still significantly less than that between instructed lie and truth conditions, which indicated that under the chosen condition, the lying process was facilitated.We can make these conclusions on the basis of these three experiments:there are differences in at least three cognitive components between the lying and truth-telling reaction; we can produce cognitive conflicts after we only realize that we intend to lie next; there are difference between the cognitive process of choosing reaction and that of instructed reaction. Under the condition of autonomous choice, lying is facilitated.