Proof of Theorem 1 — the difference between Bayesian and DeGroot cutoffs
ProvedPostTruth.Regimes.threshold_gapmisinformationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1social-learning
For and , let be the two-narrative gap and let be the DeGroot cutoff when the true state is . Then
The right-hand expression is strictly decreasing as a function of positive . This identity locates the change in the ordering of the two societies' mislearning cutoffs.
Preamble
import Mathlib import Definitions.Def_PostTruth_Regimes_Model
Formal statement
namespace PostTruth.Regimes
/-- The displayed cutoff difference in the proof of Theorem 1, p. 20. -/
theorem threshold_gap (p q : ℝ) (hpq : Standing p q) :
gap p q - rTilde p q = (2 * p * (1 - q) - 1) / (2 * q) ∧
(gap p q - rTilde p q = 0 ↔ q = qStar p) ∧
(gap p q - rTilde p q < 0 ↔ qStar p < q) ∧
StrictAntiOn (fun t : ℝ => (2 * p * (1 - t) - 1) / (2 * t))
(Set.Ioi 0) := by sorry
end PostTruth.Regimes
Source
Mostagir & Siderius, Learning in a Post-Truth World, SSRN working paper 3879174 (version of 2 July 2021), p. 20, proof of Theorem 1 (displayed cutoff difference); https://ssrn.com/abstract=3879174
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.