关于「东西方原典考辨:「巧言令色」「眸子」与亚里士多德的品格说服」,《三秒薄切片信任》给出的定量结论是什么?
结论:A 段显示,先验为一半时,若巧言令色的似然比是 0.5(补偿型),看到它之后,对方可靠的后验降到 0.333;先验为 0.8 时降到 0.667。若似然比是 1(中性),后验不变。一条可核验的交付记录,似然比取 3,后验升到 0.750 或 0.923;两条独立记录,似然比 9,后验升到 0.900 与 0.973。巧言令色加一条记录,似然比 0.5 乘 3 等于 1.5,后验 0.600。
数据口径:书中参数为示意;全部计算由书中程序生成;本书只写如何让真实、可兑现的信号在前三秒清晰一致,不写任何伪造资质或假权威的手法,不宣称任何气味的健康功效,是非个人化经营建议。 第一章区分信度与效度,示意参数下三秒判断只达到效度上限的 0.612,与长期评价的一致率为 0.579。
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。 与结论吻合的打印值: 0.333, 0.5, 0.600, 0.667, 0.750, 0.8
A. posterior P(R) after the signal signal LR prior 0.5 prior 0.8 polish, compensating 0.5 0.333 0.667 polish, neutral 1.0 0.500 0.800 verifiable record 3.0 0.750 0.923 polish (0.5) + record (3) 1.5 0.600 0.857 two independent records 9.0 0.900 0.973 B. naive reading of polish (LR 2) vs assumed truth (LR 0.5) prior 0.3: naive 0.462, warranted 0.176, overtrust +0.285 prior 0.5: naive 0.667, warranted 0.333, overtrust +0.333 prior 0.8: naive 0.889, warranted 0.667, overtrust +0.222
查看完整验算程序
# Chapter 7. p71: "fine words and an ingratiating face" as a likelihood
# ratio. Let R be "this presenter is reliable" with prior P(R). A polished
# surface S has likelihood ratio LR = P(S | R) / P(S | not R). If polish is
# cheap and those with less to offer invest more in it, LR < 1 and polish
# should lower belief, which is one reading of the Analects line. A
# verifiable record V (public returns kept, past orders delivered) is
# costly to fake, LR > 1. Signals independent given R multiply their LRs.
# A naive observer who reads polish as LR = 2 is compared with the
# posterior under the assumed true LR. All LRs are illustrative.
def post(prior, lr):
odds = prior / (1.0 - prior) * lr
return odds / (1.0 + odds)
if __name__ == "__main__":
print("A. posterior P(R) after the signal")
print(" signal LR prior 0.5 prior 0.8")
for name, lr in (("polish, compensating", 0.5),
("polish, neutral", 1.0),
("verifiable record", 3.0),
("polish (0.5) + record (3)", 1.5),
("two independent records", 9.0)):
print(" {:<27} {:.1f} {:.3f} {:.3f}".format(
name, lr, post(0.5, lr), post(0.8, lr)))
print("B. naive reading of polish (LR 2) vs assumed truth (LR 0.5)")
for prior in (0.3, 0.5, 0.8):
a, b = post(prior, 2.0), post(prior, 0.5)
print(" prior {:.1f}: naive {:.3f}, warranted {:.3f}, "
"overtrust {:+.3f}".format(prior, a, b, a - b))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《三秒薄切片信任捕获:视听嗅三通道同频一致性与瞬间权威建立母典》第 7 章「东西方原典考辨:「巧言令色」「眸子」与亚里士多德的品格说服」(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-419 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026419,
author = {Fan, Yafu},
title = {Three-Second Thin-Slicing Trust: Visual-Auditory-Olfactory Congruence and Instant Authority Signalling},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}