关于「实证闭环:形象钉扎台账、实战工具与检查器规格」,《具象词汇与形象钉扎》给出的定量结论是什么?

结论:第八章给出形象钉扎台账、蒸馏小工具「具象度与形象钉扎评分器」、按四件套分组的检查器规格与预注册检验,主检验每组需要 163 名读者。全书示意文案只用合成商品与合成人物,不点名评价任何真实品牌。书中以标注框区分已发表的证据、可运行的 Python 计算(22 段程序)、原典与作者的假说。

数据口径:具体性效应真实存在,帕维奥的双编码理论与散文层面的实验都支持它;可「倍数」取决于设计,混合呈现放大比值,纯呈现与丰富语境缩小比值,示意参数下同一批词的纯表比值 1.68、混合表比值 3.29。 书中参数为示意,全部计算由书中程序生成;本书是非个人化经营建议,不构成法律意见,不承诺任何销量或收入。

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: 163

A. readers per arm, plain p1 versus concrete p2
0.30 vs 0.40: power 0.8 ->356power 0.9 ->477
0.30 vs 0.45: power 0.8 ->163power 0.9 ->217
0.20 vs 0.35: power 0.8 ->138power 0.9 ->185
B. paired readers (McNemar), discordant p10 / p01
0.20 / 0.10: power 0.8 ->234power 0.9 ->312
0.25 / 0.10: power 0.8 ->120power 0.9 ->160
C. Monte Carlo power at n = 163 per arm: 0.789
查看完整验算程序
# Chapter 8. p82: sample size for a pre-registered free-recall test.
# Design A (between readers): half read the concrete blurb, half the plain
# one; outcome = recalls the signature feature after 7 days. Normal
# approximation for two proportions, two-sided alpha 0.05.
# Design B (within readers, McNemar): each reader sees both versions of
# two different synthetic items; p10 / p01 are the discordant shares.
# A Monte Carlo of 4,000 studies (seed fixed) checks design A's power.
import math
import random
from statistics import NormalDist

Z = NormalDist()


def n_two(p1, p2, alpha=0.05, power=0.8):
    za, zb = Z.inv_cdf(1 - alpha / 2), Z.inv_cdf(power)
    pb = (p1 + p2) / 2.0
    num = (za * math.sqrt(2 * pb * (1 - pb))
           + zb * math.sqrt(p1 * (1 - p1) + p2 * (1 - p2))) ** 2
    return math.ceil(num / (p1 - p2) ** 2)


def n_mcnemar(p10, p01, alpha=0.05, power=0.8):
    za, zb = Z.inv_cdf(1 - alpha / 2), Z.inv_cdf(power)
    pd, d = p10 + p01, p10 - p01
    return math.ceil((za * math.sqrt(pd) + zb * math.sqrt(pd - d * d)) ** 2
                     / (d * d))


def mc_power(p1, p2, n, rng, studies=4000):
    za = Z.inv_cdf(0.975)
    hits = 0
    for _ in range(studies):
        a = len([1 for _ in range(n) if rng.random() < p1])
        b = len([1 for _ in range(n) if rng.random() < p2])
        pb = (a + b) / (2.0 * n)
        se = math.sqrt(2 * pb * (1 - pb) / n)
        if se > 0 and abs(b / n - a / n) / se > za:
            hits += 1
    return hits / studies


if __name__ == "__main__":
    print("A. readers per arm, plain p1 versus concrete p2")
    for p1, p2 in ((0.30, 0.40), (0.30, 0.45), (0.20, 0.35)):
        print("  {:.2f} vs {:.2f}: power 0.8 -> {:>4}   power 0.9 -> {:>4}"
              .format(p1, p2, n_two(p1, p2), n_two(p1, p2, power=0.9)))
    print("B. paired readers (McNemar), discordant p10 / p01")
    for p10, p01 in ((0.20, 0.10), (0.25, 0.10)):
        print("  {:.2f} / {:.2f}: power 0.8 -> {:>4}   power 0.9 -> {:>4}"
              .format(p10, p01, n_mcnemar(p10, p01),
                      n_mcnemar(p10, p01, power=0.9)))
    rng = random.Random(39382)
    n = n_two(0.30, 0.45)
    print("C. Monte Carlo power at n = {} per arm: {:.3f}".format(
        n, mc_power(0.30, 0.45, n, rng)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《具象词汇双编码认知、多感官意象钉扎与文字固化品牌形象母典》第 8 章「实证闭环:形象钉扎台账、实战工具与检查器规格」(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-393 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026393,
  author    = {Fan, Yafu},
  title     = {Dual-Coding Concrete Lexicons, Multi-Sensory Mental Imagery Pinning and Textual Brand-Persona Crystallisation},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论