关于「东西方原典考辨:知行、内外与心中之念」,《心智模拟与行为兑现》给出的定量结论是什么?

结论:一种每次以 0.05 概率显现的倾向,一次机会里只有 0.050 的显现概率,五次机会 0.226,二十次 0.642;每次概率为 0.50 的倾向,五次机会就有 0.969。估计精度方面,若真实守约率为七成,观察 5 次后的 95 % 区间宽 0.598,10 次后 0.500,20 次后 0.376,50 次后 0.248,100 次后仍有 0.178。所谓「形于外」,需要很多次真实行为才能「形」得清楚。

数据口径:书中参数为示意;全部计算由书中程序生成;本书不是心理、临床或婚恋咨询,也不是个人化经营建议。 第一章用学习规则说明,完全可预期的想象学会后不再产生正向误差,真正的问题是缺少负向误差:示意轨迹被模拟抬高到比真实值高出 0.600,且永不下修。

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: 0.05, 0.050, 0.178, 0.226, 0.248, 0.376

A. posterior 95 % interval for p after k occasions (70 % kept)
k kept low high width
00 0.026 0.976 0.950
54 0.360 0.958 0.598
107 0.392 0.892 0.500
2014 0.480 0.856 0.376
5035 0.562 0.810 0.248
10070 0.604 0.782 0.178
B. chance that a disposition shows at least once in k occasions
p0.05: k=1 0.050 k=5 0.226 k=20 0.642
p0.20: k=1 0.200 k=5 0.672 k=20 0.988
p0.50: k=1 0.500 k=5 0.969 k=20 1.000
查看完整验算程序
# Chapter 7. p72: "sincere within, visible without" as an inference.
# An inner disposition p (chance of a kept act per occasion) is not
# observed; only acts are. With a flat Beta(1, 1) prior and x kept acts
# in k occasions the posterior is Beta(1 + x, 1 + k - x). Its 95 %
# interval is read off a 4,000-cell midpoint grid (fsum for the mass).
# Part B: chance that a disposition p shows at least one act in k
# occasions, 1 - (1 - p)^k. Illustrative true rate 0.7.
import math


def interval(a, b, cells=4000):
    xs = [(i + 0.5) / cells for i in range(cells)]
    logs = [(a - 1) * math.log(x) + (b - 1) * math.log(1 - x)
            for x in xs]
    top = max(logs)
    w = [math.exp(v - top) for v in logs]
    tot = math.fsum(w)
    lo = hi = None
    run = []
    for x, wi in zip(xs, w):
        run.append(wi)
        c = math.fsum(run) / tot if len(run) % 8 == 0 else None
        if c is None:
            continue
        if lo is None and c >= 0.025:
            lo = x
        if hi is None and c >= 0.975:
            hi = x
    return lo, hi


def main():
    print("A. posterior 95 % interval for p after k occasions (70 %"
          " kept)")
    print("   k     kept   low     high    width")
    for k in (0, 5, 10, 20, 50, 100):
        x = round(0.7 * k)
        lo, hi = interval(1 + x, 1 + k - x)
        print("   {:>3}   {:>4}   {:.3f}   {:.3f}   {:.3f}".format(
            k, x, lo, hi, hi - lo))
    print("B. chance that a disposition shows at least once in k"
          " occasions")
    for p in (0.05, 0.2, 0.5):
        cells = ["{:.3f}".format(1 - (1 - p) ** k) for k in (1, 5, 20)]
        print("   p = {:.2f}:  k=1 {}   k=5 {}   k=20 {}".format(
            p, *cells))


if __name__ == "__main__":
    main()

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《心智模拟亲密与具身行为兑现的神经分野:多巴胺投射幻象与真实契约代价母典》第 7 章「东西方原典考辨:知行、内外与心中之念」(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-430 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026430,
  author    = {Fan, Yafu},
  title     = {Mental Simulation versus Behavioural Enactment in Intimacy: Dopaminergic Fantasy Loops, Oxytocin Bonding Costs and Commitment Asymmetries},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论