Adequacy of the Sequential-Task Paradigm in Evoking Ego-Depletion and How to Improve Detection of Ego-Depleting Phenomena

Self-control is defined as individuals' capacity to alter, modify, change, or override impulses, desires, and habitual responses (Baumeister, 2002; Muraven et al., 2005). Capacity for self-control is important and adaptive. Without it, we would be " slaves " to habits and impulses and unable to engage in sustained, goal-directed behavior. Loss of self-control has been shown to be related to numerous maladaptive health, social, and economic outcomes (Baumeister, 2002). Contemporary theories indicate that human capacity for self-control is limited (Baumeister et al., 1998). According to the strength model of self-control, performance on tasks requiring self-control draws energy from a general, unitary, and limited " internal " resource (Baumeister et al., 1998; Muraven et al., 1998). Because this resource is finite, the model predicts that engaging in tasks requiring self-control would lead to the depletion of the resource and reduced performance on subsequent self-control tasks. The state of self-control resource depletion is termed " ego-depletion. " Baumeister and colleagues tested the strength model of self-control using a " sequential-task " paradigm in which participants engaged in two consecutive tasks. For participants allocated to an experimental (ego-depletion) group, both tasks required self-control, while only the second task required self-control for participants in the control (no depletion) group (Baumeister et al., 2007). Reduced performance on the second task in participants assigned to the ego-depletion group compared to participants in the control group constituted support for the ego-depletion effect. Examples of self-control tasks typically used in sequential-task paradigm experiments include downregulating emotions while watching emotionally-arousing video-clips (e.g., Muraven et al., 1998) or suppressing an unwanted thought by attending to a distracter (e.g., Muraven et al., 2006). An initial meta-analysis revealed a medium effect size (d = 0.62) across 198 tests of the ego-depletion effect using the sequential-task paradigm (Hagger et al., 2010). Recent analyses have suggested that Hagger et al.'s meta-analytic evidence for the ego-depletion effect might have been inflated due to publication bias. Re-analyses of previous findings (Carter and McCullough, 2013, 2014) and a new meta-analysis including unpublished studies (Carter et al., 2015) revealed substantial " small study " bias in the effect size reported in Hagger et al.'s (2010) original meta-analysis. There have also been studies that failed to replicate the ego-depletion effect (e.g., Xu et al., 2014), or found it to be substantially smaller in size than reported in meta-analytic syntheses (e.g., Tuk et al., 2015). Carter et al. (2015) concluded …