关于「气味通路与普鲁斯特效应:旋钮而非开关」,《声香签名与气味记忆》给出的定量结论是什么?

结论:嗅觉初级通路不经丘脑中继,这一点有解剖依据;气味唤起的记忆情绪更浓、久未触碰,这一点有实验依据。可从这里跳到「闻到就会瞬间改变状态」,中间缺了两环:效应是平移而不是开关,效应落在情绪强度而不是准确度。程序 p11 说明,要让多数人都越过对照组的高位,所需效应量远超常见实验;程序 p12 说明,情绪差清晰时,准确度差可以完全为零。

数据口径:书中参数为示意,全部计算由书中程序生成。 本书不是医疗、心理咨询或法规意见,不宣称任何身心功效。

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

A. one shifted distribution, four readings of it
d U3 P(sup) overlap above v-P90
0.000.500 0.500 1.000 0.100
0.200.579 0.556 0.920 0.140
0.350.637 0.598 0.861 0.176
0.500.691 0.638 0.803 0.217
0.800.788 0.714 0.689 0.315
1.200.885 0.802 0.549 0.468
B. a true switch would need nearly all odour ratings above
the visual P90; that requires d of about
share 0.50 above v-P90 -> d1.28
share 0.80 above v-P90 -> d2.12
share 0.95 above v-P90 -> d2.93
查看完整验算程序
# Chapter 1. p11: a cue effect is a shifted distribution, not a switch.
# Model: ratings under a visual cue ~ N(0, 1); under an odour cue the
# whole distribution shifts by d (standardised mean difference). For
# each d we report Cohen's U3 (share of odour ratings above the visual
# median), the probability of superiority Phi(d / sqrt 2), the overlap
# of the two densities 2 Phi(-d / 2), and the share of odour ratings
# above the visual 90th percentile. All d values are illustrative.
import math
from statistics import NormalDist

N = NormalDist()
Z90 = N.inv_cdf(0.90)


def row(d):
    u3 = N.cdf(d)
    sup = N.cdf(d / math.sqrt(2.0))
    ovl = 2.0 * N.cdf(-abs(d) / 2.0)
    top = 1.0 - N.cdf(Z90 - d)
    return u3, sup, ovl, top


if __name__ == "__main__":
    print("A. one shifted distribution, four readings of it")
    print("     d     U3   P(sup)  overlap  above v-P90")
    for d in (0.0, 0.2, 0.35, 0.5, 0.8, 1.2):
        u3, sup, ovl, top = row(d)
        print("  {:>4.2f}  {:.3f}  {:.3f}   {:.3f}    {:.3f}".format(
            d, u3, sup, ovl, top))
    print("B. a true switch would need nearly all odour ratings above")
    print("   the visual P90; that requires d of about")
    for share in (0.5, 0.8, 0.95):
        d_need = Z90 - N.inv_cdf(1.0 - share)
        print("   share {:.2f} above v-P90 -> d = {:.2f}".format(
            share, d_need))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《声学节律与嗅觉分子跨模态耦合、普鲁斯特记忆唤醒与瞬时状态锚定母典》第 1 章「气味通路与普鲁斯特效应:旋钮而非开关」(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-392 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026392,
  author    = {Fan, Yafu},
  title     = {Cross-Modal Coupling of Acoustic Rhythms and Olfactory Cues, Proustian Memory Retrieval and Rapid State-Conditioning Anchors},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论