Changing a probability when evidence arrives

Reasoning under uncertainty involves revision. We begin with what is already known, consider new evidence, and change our view by an amount that the evidence justifies.

This is the basic idea behind Bayesian updating. The name refers to a mathematical rule, but the central question is understandable without a formula: “Would this evidence be more expected if one explanation were true than if the other were true?”

Evidence matters most when it helps distinguish the possibilities. This lesson shows how to combine that distinction with a sensible starting point.