关于「孤立效应与最佳反差:标志特征怎样从背景里跳出来」,《具象词汇与形象钉扎》给出的定量结论是什么?

结论:第六章模拟冯·雷斯托夫孤立效应,最优反差中位 0.43,区间 0.25 到 0.70。

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: 0.25, 0.43, 0.70

central case b 2.2, c 0.35, k 0.5: best x0.43
best x quantiles 5/25/50/75/950.25 0.35 0.43 0.56 0.70
share of draws with best x at the 1.00 edge0.000
查看完整验算程序
# Chapter 6. p62: the best contrast of a signature feature is an interval.
# Value of contrast x for a persona card: isolate gain lg(a + b x) - lg(a),
# minus the background drain 11 (lg(a) - lg(a - c x)) / 11 averaged per
# item, minus a tone penalty k x^2 (a loud feature breaks the voice).
# Grid x = 0.00 to 1.00 by 0.01. 2,000 illustrative parameter draws:
# b ~ U(1.5, 3.0), c ~ U(0.1, 0.5), k ~ U(0.2, 0.8); a = -0.4 fixed.
import math
import random

A = -0.4


def lg(z):
    return 1.0 / (1.0 + math.exp(-z))


def value(x, b, c, k):
    return (lg(A + b * x) - lg(A)) - (lg(A) - lg(A - c * x)) - k * x * x


def best_x(b, c, k):
    grid = [i / 100.0 for i in range(101)]
    return max(grid, key=lambda x: value(x, b, c, k))


if __name__ == "__main__":
    print("central case b 2.2, c 0.35, k 0.5: best x = {:.2f}"
          .format(best_x(2.2, 0.35, 0.5)))
    rng = random.Random(39362)
    xs = []
    for _ in range(2000):
        b = rng.uniform(1.5, 3.0)
        c = rng.uniform(0.1, 0.5)
        k = rng.uniform(0.2, 0.8)
        xs.append(best_x(b, c, k))
    xs.sort()
    q = [xs[int(p * len(xs))] for p in (0.05, 0.25, 0.5, 0.75, 0.95)]
    print("best x quantiles 5/25/50/75/95: " + " ".join(
        "{:.2f}".format(v) for v in q))
    print("share of draws with best x at the 1.00 edge: {:.3f}".format(
        len([1 for v in xs if v >= 1.0]) / len(xs)))

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 6 章「孤立效应与最佳反差:标志特征怎样从背景里跳出来」(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