关于「希克与菲茨:折叠菜单与线性路径的操作时间账」,《客群分化界面与同价同条款》给出的定量结论是什么?

结论:第二章算希克与菲茨的账:分组并不减少信息,折叠的代价在于发现,对偏好清晰版的访客,零折叠线性路径多走完 +0.3076 的旅程。

数据口径:书中参数为示意,全部计算由书中程序生成,不对应任何真实站点;非法律建议,非个人化经营建议。 第五章说明选择过载是有条件的效应:示意合并均值接近零,新情境的预测区间却从 -0.5195 到 +0.6687;筛选加买家自选排序让比较楼盘快 5.69 倍。

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: 0.3076

runs per cell: 20,000
groupFOLD doneFOLD sLINEAR doneLINEAR sgain
elder-pref0.691011.570.99866.12+0.3076
fashion0.95756.401.00004.10+0.0425
property0.91977.631.00004.59+0.0803
narrative0.86998.551.00004.92+0.1301
查看完整验算程序
# Chapter 2. p23: a four-step journey, folded menus versus a zero-fold
# linear path, seeded Monte Carlo (illustrative parameters throughout).
# Each step = one Hick decision + one Fitts tap (p21, p22 models).
# FOLD: 24 px menu icon at step 1 (not recognised with probability m,
# then 6.0 s search), 32 px items, hover-only submenus that fail on
# touch with probability h (a dead end, 2.5 s lost). LINEAR: every step
# shows at most 4 visible choices of 44 px. A frustration event (miss,
# dead end, unfound icon) makes the visitor leave with probability q.
import math
import random

GROUPS = [  # name, b_hick, b_fitts, sd px, m, h, q
    ("elder-pref", 0.27, 0.16, 7.0, 0.25, 0.30, 0.22),
    ("fashion", 0.15, 0.10, 4.0, 0.05, 0.10, 0.10),
    ("property", 0.17, 0.12, 5.0, 0.08, 0.15, 0.12),
    ("narrative", 0.19, 0.13, 5.5, 0.10, 0.20, 0.15),
]
D = 300.0


def tap(rng, b, sd, w):
    t, events = 0.0, 0
    while True:
        t += 0.10 + b * math.log2(2 * D / w)
        if abs(rng.gauss(0.0, sd)) <= w / 2.0:
            return t, events
        events += 1


def journey(rng, g, fold):
    _, bh, bf, sd, m, h, q = g
    t = 0.0
    for step in range(4):
        ev = 0
        if fold:
            if step == 0:
                dt, e = tap(rng, bf, sd, 24.0)
                t += 0.20 + bh + dt
                ev += e
                if rng.random() < m:
                    t += 6.0
                    ev += 1
            if rng.random() < h:
                t += 2.5
                ev += 1
            dt, e = tap(rng, bf, sd, 32.0)
            t += 0.20 + bh * math.log2(5) + dt
        else:
            dt, e = tap(rng, bf, sd, 44.0)
            t += 0.20 + bh * math.log2(5) + dt
        ev += e
        for _ in range(ev):
            if rng.random() < q:
                return False, t
    return True, t


rng = random.Random(20261120)
RUNS = 20000
print("runs per cell: {:,}".format(RUNS))
print("group        FOLD done  FOLD s   LINEAR done  LINEAR s  gain")
for g in GROUPS:
    res = {}
    for fold in (True, False):
        ok, times = 0, []
        for _ in range(RUNS):
            done, t = journey(rng, g, fold)
            if done:
                ok += 1
                times.append(t)
        res[fold] = (ok / RUNS, math.fsum(times) / len(times))
    f, li = res[True], res[False]
    print("{:<11}   {:.4f}   {:>5.2f}     {:.4f}     {:>5.2f}   {:+.4f}"
          .format(g[0], f[0], f[1], li[0], li[1], li[0] - f[0]))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《客群认知画像分层、适老与新世代界面分化及千人千面转化交互母典》第 2 章「希克与菲茨:折叠菜单与线性路径的操作时间账」(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-363 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026363,
  author    = {Fan, Yafu},
  title     = {Demographic Cognitive Segmentation, Elder-to-Gen-Z Interface Divergence and Audience-Adaptive Conversion UX},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论