23 Labs Tried to Replicate Ego Depletion, and It Didn't Show Up

Rather than one lab running one more version of the classic experiment, this project coordinated 23 separate laboratories around the world to run the exact same, pre-agreed protocol on over 2,100 participants combined, with the analysis plan locked in before any data was collected, precisely to prevent the kind of selective reporting that had clouded the earlier literature. The result: pooling every lab's data, the ego-depletion effect came out to essentially zero, with a confidence interval comfortably straddling both positive and negative values, meaning the data were equally consistent with a genuine small effect, no effect at all, or even a small effect in the opposite direction.

Being a coordinated multi-lab project is exactly what gives this result its weight, 23 independent teams following identical, pre-registered procedures removes many of the ways a single lab's result can be shaped, consciously or not, by researcher choices made after seeing the data. It doesn't prove that no version of mental depletion exists, only that this specific, foundational experimental paradigm, the one the whole ego-depletion literature was built on, doesn't reliably produce the effect it was designed to detect.