After a pair of self-control-intensive tasks, sucrose swishing improves subsequent working memory performance

The limited strength model of self-control predicts that acts of self-control impair subsequent performance on tasks that require self-control (i.e., "ego depletion"), and the majority of the published research on this topic is supportive of this prediction. Additional research suggests that this effect can be alleviated by manipulating participants' motivation to perform-for instance, by having participants swish a drink containing carbohydrates, which is thought to function as a reward-or by requiring participants to complete two initial acts of self-control rather than only one.

Here, we explore both the effect of having participants perform two initial tasks thought to require self-control (versus two less self-control-intensive tasks) and the effect of swishing a drink containing sucrose (compared to control drinks) on subsequent self-control. Outcomes were analyzed using standard null hypothesis significance testing techniques (e.g., analysis of variance, t-tests). In some cases, test statistics were transformed into Bayes factors to aid in interpretation (i.e., to allow for acceptance of the null hypothesis).

We found that performing two self-control-intensive tasks actually improved subsequent self-control when participants swished a drink containing sucrose between tasks. For participants who swished control drinks, we found no evidence of ego depletion.

We conclude that claims that self-control failure is caused by the depletion of a resource (or that it functions as if it relies on a limited resource) merit greater circumspection. Our results-all of which were either null or contrary to predictions from the limited strength model-are important for researchers interested in patterns of self-control failure.