Liar liar neurons fire: how executive control processes contribute to the ability to deceive

摘要

This thesis presents a series of empirical investigations into the executive demands of deception. The first two experiments investigated whether the executive demands of deception are sufficient to influence receiver perceptions of credibility. Participant-senders in Study 1 (n = 52) and Study 2 (n = 97) completed a false opinion task and a battery of cognitive tasks. Deception performance was operationalized via participant-receiver judgements of veracity (Study 1, n = 624; Study 2, n = 1140). While the results from Study 1 showed a small positive relationship between executive abilities and deception performance, the results from Study 2 were stronger. They indicated that while working memory skill had a moderate positive relationship with deception performance, set shifting and inhibitory control skills were unrelated to deception performance once working memory skill had been taken into account. The third study used a resource depletion framework to experimentally manipulate executive abilities. Participant-senders (n = 114) completed two false opinion tasks; one before the administration of a cognitive task (either an executive task designed to deplete the availability of executive resources or one of two control tasks) and the other immediately after. Once again deception performance was operationalized via participant-receiver judgements of veracity (n = 798). The results indicated that while deception performance was impaired by the executive task, it was relatively unaffected by either of the control tasks. The fourth study presents a theoretical analysis assessing the appropriateness of standard by-judge and by-sender aggregating procedures commonly used in deception detection research. A series of Monte Carlo simulations demonstrated that the aggregation of deception data can cause inflated Type 1 error rates and poor statistical power and that Generalized Linear Mixed Models (GLMMs) may overcome these problems. Consequently, a series of GLMMs were used to re-analyze the data from Study 3. The results were consistent with previous analyses. Overall, the evidence reported in this thesis demonstrates that the demands of deceiving in false opinion tasks are sufficient to influence a person’s behaviours such that those with poor executive abilities tend to be worse liars than those with good executive abilities.