关于「偏好带:中等分形维数的证据、样本与漂移」,《亲生命分形界面与首屏信任》给出的定量结论是什么?

结论:第二章用自助法说明 24 名被试只能把峰位钉在约 0.098 宽的区间里,测试图片中精确分形占一半时合并峰位移到 1.55。

数据口径:核对原文后,本书写下蓝图修正:中等维数受偏爱,但证据来自小样本实验室静态图片,峰位随刺激类型与人群漂移;「降压」只是皮肤电导、脑电 alpha 一类实验室压力标志物的差异,不是血压读数,本书不宣称任何健康功效;「显著提升转化」改写为带推翻条件的中介链假说;「黄金配比」改写为用自家数据拟合的最优区间。 书中参数为示意;全部计算由书中程序生成;本书不是医疗建议,也不是个人化经营建议。

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: 0.098, 24

true peak 1.40; quadratic fit; 1,000 bootstrap resamples
subjectspeaklo95hi95width
121.3771.2531.4370.184
241.4081.3521.4490.098
481.3661.2901.4060.116
961.3891.3451.4190.074
band 1.3 to 1.5 has width 0.2
查看完整验算程序
# Chapter 2. p21: a noisy quadratic preference curve and its peak interval
# Synthetic lab study: each subject rates nine images, D = 1.1 to 1.9.
# Rating = 5 - 6 (D - 1.4)**2 + subject offset + noise (illustrative,
# shaped after the mid-D peaks reported in the literature). A quadratic
# is fitted to the pooled ratings and the peak -b1 / (2 b2) is
# bootstrapped over subjects (1,000 resamples) for several sample sizes.
import math
import random
import statistics

LEVELS = [1.1 + 0.1 * i for i in range(9)]


def solve3(a, b):
    m = [row[:] + [v] for row, v in zip(a, b)]
    for c in range(3):
        p = max(range(c, 3), key=lambda r: abs(m[r][c]))
        m[c], m[p] = m[p], m[c]
        for r in range(3):
            if r != c:
                f = m[r][c] / m[c][c]
                m[r] = [x - f * y for x, y in zip(m[r], m[c])]
    return [m[i][3] / m[i][i] for i in range(3)]


def peak(subjects):
    xs = [d for s in subjects for d, _ in s]
    ys = [r for s in subjects for _, r in s]
    cols = [[1.0] * len(xs), xs, [x * x for x in xs]]
    a = [[math.fsum(u * v for u, v in zip(ci, cj)) for cj in cols]
         for ci in cols]
    b = [math.fsum(u * y for u, y in zip(ci, ys)) for ci in cols]
    b0, b1, b2 = solve3(a, b)
    return -b1 / (2 * b2) if b2 < 0 else float("nan")


def make(n, rng):
    out = []
    for _ in range(n):
        off = rng.gauss(0.0, 0.8)
        out.append([(d, 5 - 6 * (d - 1.4) ** 2 + off + rng.gauss(0, 1.0))
                    for d in LEVELS])
    return out


rng = random.Random(41821)
print("true peak 1.40; quadratic fit; 1,000 bootstrap resamples")
print("{:>9} {:>8} {:>8} {:>8} {:>8}".format(
    "subjects", "peak", "lo95", "hi95", "width"))
for n in (12, 24, 48, 96):
    data = make(n, rng)
    est = peak(data)
    boots = []
    for _ in range(1000):
        boots.append(peak([data[rng.randrange(n)] for _ in range(n)]))
    boots.sort()
    lo, hi = boots[24], boots[974]
    print("{:>9} {:>8.3f} {:>8.3f} {:>8.3f} {:>8.3f}".format(
        n, est, lo, hi, hi - lo))
print("band 1.3 to 1.5 has width {:.1f}".format(1.5 - 1.3))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《亲生命自然分形与几何对称:数字界面潜意识降压与首屏信任度工程母典》第 2 章「偏好带:中等分形维数的证据、样本与漂移」(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-418 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026418,
  author    = {Fan, Yafu},
  title     = {Biophilic Botanical Fractals and Geometric Symmetry in Interfaces: Subconscious Stress Reduction and First-Impression Trust Lift},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论