关于「「永久钉入」的保持率:遗忘曲线、间隔重复与标志特征该不该换」,《具象词汇与形象钉扎》给出的定量结论是什么?
结论:第五章比较集中与分散曝光,等间隔四次第 30 天保持率 0.8175,集中四次只有 0.0856。
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。
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: 0.0856, 0.8175, 30
| plan | final h | R day 30 | R day 60 | R day 180 |
| massed 0,1,2,3 | 7.61 | 0.0856 | 0.0056 | 0.0000 |
| equal 0,7,14,21 | 30.95 | 0.8175 | 0.4175 | 0.0284 |
| expanding 0,2,6,14 | 24.55 | 0.6366 | 0.2729 | 0.0092 |
| one showing | 2.00 | 0.0000 | 0.0000 | 0.0000 |
查看完整验算程序
# Chapter 5. p52: four showings of one signature feature, three schedules.
# Exponential forgetting with half-life h. Each showing resets retention
# to 1 and stretches h by (1 + a (1 - R)), where R is the retention just
# before the showing: a showing that meets a half-faded trace helps more
# (spacing). Start h0 = 2 days, a = 3; all values illustrative.
import math
H0, A = 2.0, 3.0
PLANS = {
"massed 0,1,2,3": (0, 1, 2, 3),
"equal 0,7,14,21": (0, 7, 14, 21),
"expanding 0,2,6,14": (0, 2, 6, 14),
"one showing": (0,),
}
def retention_at(plan, day):
h, last = H0, None
for s in plan:
if last is not None:
r = math.exp(-math.log(2.0) * (s - last) / h)
h *= 1.0 + A * (1.0 - r)
last = s
return math.exp(-math.log(2.0) * (day - last) / h), h
if __name__ == "__main__":
print("plan final h R day 30 R day 60 R day 180")
for name, plan in PLANS.items():
r30, h = retention_at(plan, 30)
r60, _ = retention_at(plan, 60)
r180, _ = retention_at(plan, 180)
print("{:<20} {:>7.2f} {:.4f} {:.4f} {:.4f}"
.format(name, h, r30, r60, r180))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 5 章「「永久钉入」的保持率:遗忘曲线、间隔重复与标志特征该不该换」(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}
}