关于「原典考辨:井蛙与「知之为知之」,培根的格言四十六与苏格拉底的自知无知」,《先验消解与反逆火沟通》给出的定量结论是什么?

结论:四十题答对 23 题,比例 0.575。不做任何收回时,平均信心 0.8115,高出实际 0.2365,布里尔分数 0.2908,比对所有题都说「一半一半」的 0.2500 还差。收回一半时,平均信心 0.6966,布里尔分数 0.2475,刚好胜过一半一半。完全收回时,平均信心 0.5816,高出实际只有 0.0066,布里尔分数 0.2316。

数据口径:书中参数为示意,全部计算由书中程序生成;本书非心理咨询、不提供任何临床服务、非个人化经营建议,例子只用商品与消费认知,不涉及政治、宗教与族群议题。 第六章把追问、肯定、证据与思想实验写成马尔可夫决策过程,示意参数下最难起点的期望收益 10.555,参数扰动两千次,完整路线不变的比例为 0.8270,沿用基准策略的平均相对遗憾只有 0.0003,所以只称模拟最优。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.0066, 0.2316, 0.2365, 0.2475, 0.2500, 0.2908

questions 40, answered right 23 (0.575)
shrink smean confoverconfBrier
0.000.8115+0.23650.2908
0.250.7540+0.17900.2657
0.500.6966+0.12160.2475
0.750.6391+0.06410.2362
1.000.5816+0.00660.2316
saying 0.5 to everything: Brier 0.2500
查看完整验算程序
# Chapter 7. p72: "to know what you know": calibration before and after.
# Synthetic: 40 product questions (ingredients, price parts, customs
# rules). A respondent's chance of being right is p_i; the stated
# confidence is inflated: conf = min(0.99, p_i + 0.25). After the "explain
# how it works" step, confidence is pulled toward the truth by a factor
# s: conf' = p_i + (conf - p_i)(1 - s). Brier score = mean (conf - y)^2
# with y the seeded 0/1 outcome. Overconfidence = mean conf - share right.
# All synthetic and illustrative.
import math
import random

N_Q = 40


def brier(confs, ys):
    return math.fsum((c - y) ** 2 for c, y in zip(confs, ys)) / len(ys)


if __name__ == "__main__":
    rng = random.Random(38773)
    ps = [rng.uniform(0.3, 0.9) for _ in range(N_Q)]
    ys = [1 if rng.random() < p else 0 for p in ps]
    hit = 0
    for y in ys:
        hit += y
    print("questions {}, answered right {} ({:.3f})".format(
        N_Q, hit, hit / N_Q))
    base = [min(0.99, p + 0.25) for p in ps]
    print("  shrink s   mean conf   overconf   Brier")
    for s in (0.0, 0.25, 0.5, 0.75, 1.0):
        cs = [p + (c - p) * (1 - s) for p, c in zip(ps, base)]
        mc = math.fsum(cs) / N_Q
        print("   {:5.2f}     {:.4f}     {:+.4f}   {:.4f}".format(
            s, mc, mc - hit / N_Q, brier(cs, ys)))
    flat = [0.5] * N_Q
    print("  saying 0.5 to everything: Brier {:.4f}".format(
        brier(flat, ys)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《顽固先验偏见消解、解释深度错觉破除与反逆火效应沟通母典》第 7 章「原典考辨:井蛙与「知之为知之」,培根的格言四十六与苏格拉底的自知无知」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-387 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026387,
  author    = {Fan, Yafu},
  title     = {Dissolving Entrenched Priors, Shattering the Illusion of Explanatory Depth and Anti-Backfire Persuasion Mechanics},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论