Module 10 · The science in depth · Evidence mixed

When famous findings shrink

Ego depletion, the marshmallow test and nudging — what the debates teach.

7 min read · Level: Advanced · Free — no account needed to read

The problem it solves

What should you do when a famous finding turns out to be smaller than claimed?

The concept

Psychology has re-tested many well-known findings, and some shrank. Ego depletion — willpower as a draining resource — showed an effect close to zero in a 23-lab replication. A larger, more varied version of the marshmallow test found a much weaker link between childhood delay and later success than the originals.

Nudging is debated too: one meta-analysis found a small-to-medium average effect; a re-analysis adjusting for unpublished results found little average effect, while suggesting some kinds may still work.

This isn’t a reason to distrust all research. It’s a reason to prefer findings that have been repeated, to look for effect sizes, and to treat any single famous study as a starting point.

Why it matters

Habit advice is full of confident claims built on single studies. Knowing how findings can shrink helps you weigh advice — and is why this lab labels evidence as mixed where it is.

Evidence mixed: Some studies support it and others do not, or early findings shrank when researchers tried to repeat them. The lesson explains both sides.

See it happen

A productivity blog says you should do hard tasks first because “willpower runs out by the afternoon — science proves it”.

The ordinary first approach

If science proves it, it must be true.

With the idea

The reader knows the willpower finding is disputed, and tests hard-tasks-first on themselves as a personal experiment.

What changed

A “proven” rule became a hypothesis worth testing.

Easy to get wrong

• Believing a finding because it’s famous.
• Dismissing all of psychology because some findings shrank.
• Ignoring whether a claim has been replicated.

Different perspectives

Where the field genuinely disagrees. None of these is presented as the answer.

Replication failures show the system working

Re-testing and correcting is how science improves; shrinking effects mean the field is taking evidence seriously.

Some effects may be real but fragile

Researchers still argue that certain versions of these effects exist under specific conditions, and that a failed replication tests one version.

Try it now: Weigh a claim

1. Take one habit claim you’ve heard presented as science.
2. Ask: has it been replicated? How big is the effect? In whom?
3. Decide whether to rely on it or test it on yourself.

Apply it to one of your habits

Is there a “scientific” rule you apply to one of your habits that deserves testing rather than trusting?

Reflect

• Which habit claim have you trusted mainly because it’s famous?

Go deeper — evidence & sources

• Hagger, Chatzisarantis, Alberts, Anggono, Batailler, Birt and colleagues (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science. doi:10.1177/1745691616652873
• Baumeister, Bratslavsky, Muraven & Tice (1998). Ego depletion: Is the active self a limited resource?. Journal of Personality and Social Psychology. doi:10.1037/0022-3514.74.5.1252
• Watts, Duncan & Quan (2018). Revisiting the Marshmallow Test: A Conceptual Replication Investigating Links Between Early Delay of Gratification and Later Outcomes. Psychological Science. doi:10.1177/0956797618761661
• Mertens, Herberz, Hahnel & Brosch (2022). The effectiveness of nudging: A meta-analysis of choice architecture interventions across behavioral domains. Proceedings of the National Academy of Sciences. doi:10.1073/pnas.2107346118
• Maier, Bartoš, Stanley, Shanks, Harris & Wagenmakers (2022). No evidence for nudging after adjusting for publication bias. Proceedings of the National Academy of Sciences. doi:10.1073/pnas.2200300119