关于「液压与复演:两种宣泄模型的保序检验」,《宣泄证伪与冷却期》给出的定量结论是什么?

结论:第二章以保序回归比较液压模型与复演模型,愤怒上的似然比为 164.5,并给出宣泄可能看起来有效的条件与反方证据。

数据口径:书中参数为示意;全部计算由书中程序生成;本书不是心理、医疗或法律意见,也不是个人化经营建议。 第四章用 Zillmann 的两只时钟计算兴奋转移窗口,示意峰值在 10.75 分钟,并并陈双因素理论的争议。

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: 164.5

measuremodelfitted means (cont, dist, rumi)chi2
angerrehearsal26.250,27.320,29.7800.00
angerhydraulic27.714,27.714,27.71410.21
angerlikelihood ratio rehearsal : hydraulic = 164.5
aggressionrehearsal-0.210,0.010,0.2100.00
aggressionhydraulic-0.023,-0.023,-0.0239.03
aggressionlikelihood ratio rehearsal : hydraulic = 91.2
查看完整验算程序
# Chapter 2. p21: Hydraulic versus rehearsal model, order-restricted fit.
# 液压模型:宣泄越投入,剩余愤怒与攻击越低,预测
# 对照 >= 分心 >= 反刍;复演模型:宣泄越投入、越想着冒犯者,
# 攻击越高,预测 对照 <= 分心 <= 反刍。用加权保序回归(PAVA)
# 在各自的次序约束下拟合 Bushman 2002 表 1 的三组均值(n = 200),
# 比较卡方与似然比。数据出处同 p11。
import math

N = 200
DATA = {
    "anger": ((26.25, 10.98), (27.32, 10.88), (29.78, 11.56)),
    "aggression": ((-0.21, 1.27), (0.01, 1.39), (0.21, 1.54)),
}


def pava(y, w):
    """Weighted least squares fit, non-decreasing in index order."""
    blocks = [[y[i], w[i], 1] for i in range(len(y))]
    i = 0
    while i < len(blocks) - 1:
        if blocks[i][0] > blocks[i + 1][0]:
            a, b = blocks[i], blocks[i + 1]
            wt = a[1] + b[1]
            m = math.fsum([a[0] * a[1], b[0] * b[1]]) / wt
            blocks[i:i + 2] = [[m, wt, a[2] + b[2]]]
            i = max(i - 1, 0)
        else:
            i += 1
    fit = []
    for m, _, k in blocks:
        fit += [m] * k
    return fit


def chi2(y, w, fit):
    return math.fsum(wi * (yi - fi) ** 2 for yi, wi, fi in zip(y, w, fit))


def main():
    print("measure     model       fitted means (cont, dist, rumi)  chi2")
    for meas, rows in DATA.items():
        y = [m for m, _ in rows]
        w = [N / s ** 2 for _, s in rows]
        reh = pava(y, w)
        hyd = [-v for v in pava([-v for v in y], w)]
        c_r, c_h = chi2(y, w, reh), chi2(y, w, hyd)
        for name, fit, c in (("rehearsal", reh, c_r),
                             ("hydraulic", hyd, c_h)):
            f = ", ".join("%7.3f" % v for v in fit)
            print("%-11s %-10s  %-31s %6.2f" % (meas, name, f, c))
        lr = math.exp((c_h - c_r) / 2)
        print("%-11s likelihood ratio rehearsal : hydraulic = %.1f"
              % (meas, lr))


if __name__ == "__main__":
    main()

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《愤怒宣泄假说证伪、情绪唤醒双因素衰减与谈判冷却期最优长度母典》第 2 章「液压与复演:两种宣泄模型的保序检验」(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-200 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026200,
  author    = {Fan, Yafu},
  title     = {Falsifying the Catharsis Hypothesis: Two-Factor Arousal Decay and the Optimal Cooling-Off Period in Negotiation},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论