Skip to content

Belief Updating & Cognitive Bias

Primes about how beliefs shift, or fail to, under evidence and social exposure: Bayesian updating and Cromwell's rule as normative baselines, contrasted with biases that distort updating — anchoring, optimism bias, mere exposure, and the false-consensus-like traps of unanimity and pluralistic ignorance.

8 primes in this family — primes that sit near one another in abstraction space (k-means over structural-signature embeddings). Each is shown with its short description.

  • Anchoring — Overweight initial info.
  • Bayesian Updating — Update beliefs with evidence.
  • Cromwell's Rule — Never assign prior probability of exactly 0 or 1 to a contingent proposition, because multiplicative Bayesian updating can never move a belief away from those endpoints.
  • Lindy Effect — For entities that do not age, the longer they have already survived, the longer their expected remaining survival becomes.
  • Mere Exposure Effect — Repeated exposure to a stimulus raises liking for it along a saturating curve, even without conscious recognition of the prior encounters.
  • Optimism Bias — Overestimate positive outcomes.
  • Paradox of Unanimity — Agreement beyond what independent observation could produce becomes evidence against the conclusion, signaling a broken-independence failure mode.
  • Pluralistic Ignorance — A shared false second-order belief in which most members of a group privately reject a position yet, misreading each other's public conformity as private conviction, each believes themselves a rare exception — so the rejected position survives as an apparent consensus no one actually holds.