Are implicit attitudes toward dishonesty associated with self-serving dishonesty? Implications for the reliability of the IAT

Experiments assessing the prevalence and magnitude of dishonesty have provided a large body of empirical findings regarding the cognitive nature of honesty. However, the personal factors that regulate dishonest behavior have yet to be fully clarified. This study examined two factors that potentially inhibit dishonesty—implicit attitudes toward dishonesty and executive control. In Study 1, the participants completed the Implicit Association Test (IAT), which measured their implicit attitudes toward dishonesty, and a working memory (WM) task, which was used to index executive control. The participants subsequently completed an incentivized coin-flip prediction task wherein they were given real and repeated opportunities for dishonest reward acquisition and punishment avoidance. The results revealed that individuals showing stronger negative implicit attitudes toward dishonesty engaged in a lower frequency of dishonest behavior for punishment avoidance, although this effect was marginal. In contrast, WM capacity was not associated with variations in dishonest reward acquisition and punishment avoidance. A follow-up experiment on other-serving dishonesty, where dishonest reward acquisition and punishment avoidance were credited to two other anonymous participants, revealed that neither implicit attitudes toward dishonesty nor WM capacity was associated with dishonest behavior. An additional preregistered experiment in Study 2 demonstrated that the association between implicit attitudes toward dishonesty and self-serving dishonesty for punishment avoidance was again marginal. While it is tempting to conclude that implicit attitudes toward dishonesty are associated with self-serving dishonesty, the present study provides only weak evidence that should be interpreted with great caution. Implications for the reliability of the IAT are discussed. •Participants engaged in the IAT and WM and cognitive tasks to measure dishonesty.•The association between IAT D-scores and self-serving dishonesty was weak.•The WM score was not reliably associated with self-serving dishonesty.•Neither IAT D-scores nor WM scores were associated with other-serving dishonesty.•Implications for the reliability of the IAT are discussed.