Big Think Think more rationally with Bayes’ rule | Steven Pinker
Oct 14, 2025
In this discussion, Steven Pinker, a renowned Harvard psychologist and author, delves into the intricacies of Bayesian thinking. He breaks down Bayes' theorem, illustrating how it helps update beliefs based on evidence. Pinker emphasizes the importance of considering prior probabilities and the likelihood of evidence when forming judgments. However, he also warns of the limitations, especially in contexts requiring moral and fairness considerations. The conversation combines insights from psychology and practical reasoning, making complex ideas accessible and engaging.
AI Snips
Chapters
Transcript
Episode notes
Belief As A Probabilistic Degree
- Bayes' thinking lets you assign degrees of belief rather than binary judgments.
- It calibrates belief to evidence strength and prior knowledge.
The Three Ingredients Of Bayes' Rule
- Bayes' theorem combines a prior, a likelihood, and the overall probability of the evidence.
- The prior reflects what you believed before new evidence, and the likelihood reflects how expected the evidence is if the hypothesis is true.
Hoof Beats, Horses Not Zebras
- Pinker uses medical diagnosis as an example to show why common causes trump exotic ones.
- He cites the cliche: if you hear hoof beats, think horse, not zebra.