Decision fatigue: the depleted self vs. the replication crisis

Roy Baumeister's 1998 paper proposed that self-control draws from a limited reservoir — exert it on one task and you have less for the next. The framework, ego depletion, became one of the most-cited ideas in twentieth-century social psychology. Hundreds of papers found supporting evidence. Willpower by Baumeister and Tierney made the New York Times bestseller list. Steve Jobs's black turtleneck was retrofitted as decision-fatigue avoidance.

The framework is now contested in ways the popular literature has yet to absorb.

1. The original finding

Baumeister, Bratslavsky, Muraven, and Tice's foundational 1998 study had participants resist eating cookies, then attempt a difficult puzzle. The cookie-resisters gave up on the puzzle faster than controls. The interpretation: resisting cookies depleted a finite self-control resource (Baumeister et al., 1998).

For 15 years, the result replicated frequently. A 2010 meta-analysis of 83 studies reported a medium effect size (Hagger et al., 2010). The framework was taken as settled.

2. The 2016 multi-lab replication

Then in 2016, the Many Labs project ran a pre-registered replication of the standard ego-depletion paradigm across 23 labs and 2,141 participants. The result: essentially zero effect. The pooled effect size was 0.04 — indistinguishable from no effect at all (Hagger et al., 2016).

Subsequent meta-analyses accounting for publication bias have suggested the true effect size in the underlying literature is small or negligible (Carter et al., 2015). The 2010 meta-analysis, which had supported the effect, was retrospectively shown to be heavily inflated by publication bias.

Baumeister contested the replication results, arguing the chosen paradigm was suboptimal. The exchange continues. But the broader consensus in the field has shifted: ego depletion, as originally formulated, does not robustly replicate.

3. What might be left

This doesn't mean self-control is unlimited. It means the specific glucose-depletion / muscle-fatigue model that the original framework proposed is probably wrong.

Several alternative accounts have emerged:

Motivation shifts. What looks like depletion may be a shift in what you want to do, not a depletion of capacity. Effortful tasks are aversive; doing one makes the next aversive task feel less worth doing (Inzlicht et al., 2014).

Belief effects. Carol Dweck's lab showed that people who believe willpower is limited show depletion effects; people who believe it isn't, don't (Job et al., 2010). The construct may be partly a self-fulfilling belief.

Attention-shift. Self-control tasks engage the same systems as attention-shift tasks. Fatigue in one engages fatigue in the other for reasons that have less to do with "depletion" than with "the same neural circuit being asked to do a lot."

4. The popular implications

The "make decisions in the morning when your willpower is fresh" advice rests on the ego-depletion framework. The advice may still be useful — mornings are quieter, the prefrontal cortex is well-rested — but the theoretical justification is much shakier than 2010-era pop psychology suggested.

The "save your decisions for what matters" framing has more support, but for different reasons than originally claimed. Reducing the number of low-stakes choices is useful because it reduces cognitive load, not because it preserves a finite resource.

This is a careful distinction. The behavioral advice often survives even when its theoretical foundation collapses.