Will, habit, and the limits of character
How much of what you believe about your mind would survive a replication?
What it argues
Of one hundred replication attempts of published psychology findings, thirty-six per cent produced significant results against ninety-seven per cent in the originals, and effect sizes roughly halved. Social priming and self-report effects were the weakest. Any theory of yourself built on a single study is probably built on sand.
The probe
Before adopting a claim about your mind, check whether it replicated. If you cannot find a replication, treat it as a hypothesis about you, not a fact.
The source
Estimating the Reproducibility of Psychological Science, Science 349 — Open Science Collaboration, 2015. The link goes to the text or to a catalogue record for it, not to a summary. This one is a catalogue record: the entry gives you the question and the probe, not the argument in the author's words.