manuscript.Rmd 35.4 KB
Newer Older
1
---
linushof's avatar
linushof committed
2
3
title: 'Sampling Strategies in Decisions from Experience'
author: "Linus Hof, Thorsten Pachur, Veronika Zilker"
4
5
6
7
8
9
bibliography: sampling-strategies-in-dfe.bib
output:
  html_document:
    code_folding: hide
    toc: yes
    toc_float: yes
linushof's avatar
linushof committed
10
    number_sections: no
linushof's avatar
linushof committed
11
12
13
  pdf_document:
    toc: yes
csl: apa.csl
linushof's avatar
linushof committed
14
15
16
editor_options: 
  markdown: 
    wrap: sentence
17
18
---

19
20
```{r}
# load packages
linushof's avatar
linushof committed
21
22
pacman::p_load(repro,
               tidyverse,
linushof's avatar
linushof committed
23
24
               knitr, 
               viridis)
25
26
```

27
# Author Note
28

29
This document was created from the commit with the hash `r repro::current_hash()`. 
30

31
32
- Add information on how to reproduce the project.
- Add contact.
linushof's avatar
linushof committed
33

34
# Abstract
linushof's avatar
linushof committed
35

36
37
A probability theoretic definition of sampling and a rough stochastic model of the random process underlying decisions from experience are proposed.
It is demonstrated how the stochastic model can be used a) to explicate assumptions about the sampling and decision strategies that agents may apply and b) to derive predictions about the resulting decision behavior in terms of function forms and parameter values.
38
Synthetic choice data is simulated and modeled in cumulative prospect theory to test these predictions. 
39

linushof's avatar
linushof committed
40
# Introduction
41

42
43
...

44
## Random Processes in Sequential Sampling 
linushof's avatar
linushof committed
45

linushof's avatar
linushof committed
46
In research on the decision theory, a standard paradigm is the choice between at least two prospects.
47
48
Let a prospect be a probability space $(\Omega, \mathscr{F}, P)$.
$\Omega$ is the sample space 
linushof's avatar
linushof committed
49

50
$$\begin{equation}
linushof's avatar
linushof committed
51
\Omega = \{\omega_1, ..., \omega_n\}
52
\end{equation}$$ 
linushof's avatar
linushof committed
53

linushof's avatar
linushof committed
54
containing a finite set of possible outcomes $\omega$, monetary gains and/or losses respectively. 
55
$\mathscr{F}$ is the set of all possible subsets of $\Omega$: 
linushof's avatar
linushof committed
56

57
$$\begin{equation}
linushof's avatar
linushof committed
58
\mathscr{F} = \{A_1, A_2, ...\} = \mathscr{P}(\Omega) 
59
\; .
60
\end{equation}$$
linushof's avatar
linushof committed
61

62
$P$ is a probability mass function  
63
64

$$\begin{equation}
65
P: \mathscr{F} \mapsto [0,1] 
66
67
\end{equation}$$

linushof's avatar
linushof committed
68
that assigns each outcome $\omega$ a probability $0 < p(\omega) \leq 1$ with $P(\Omega) = 1$ [ @kolmogorovFoundationsTheoryProbability1950, pp. 2-3].
69

70
In such a choice paradigm, agents are asked to evaluate the prospects and build a preference for either one of them. 
linushof's avatar
linushof committed
71
It is common to make a rather crude distinction between two variants of this evaluation process [cf. @hertwigDescriptionexperienceGapRisky2009]. 
72
For decisions from description (DfD), agents are provided a full symbolic description of the prospects.
linushof's avatar
linushof committed
73
74
75
For decisions from experience [DfE; e.g., @hertwigDecisionsExperienceEffect2004], prospects are not described but must be explored by the means of sampling. 

To provide a formal definition of sampling in risky choice, we make use of the mathematical concept of a random variable and start by referring to a prospect as *"risky"* in the case where $p(\omega) \neq 1$ for all $\omega \in \Omega$.
76
Here, risky describes the fact that if agents would choose a prospect and any of its outcomes in $\Omega$ must occur, none of these outcomes will occur with certainty. 
linushof's avatar
linushof committed
77
It is acceptable to speak of the occurrence of $\omega$ as a realization of a random variable $X$ defined on a prospect iff the following conditions A and B are met: 
78

linushof's avatar
linushof committed
79
A) $X$ is a measurable function 
80
81

$$\begin{equation}
82
83
X: (\Omega, \mathscr{F})  \mapsto (\Omega', \mathscr{F'}) 
\; , 
84
85
\end{equation}$$

86
where $\Omega'$ is a set of real numbered values $X$ can take and $\mathscr{F'}$ is a set of subsets of $\Omega'$.
linushof's avatar
linushof committed
87
I.e., $\Omega$ maps into $\Omega'$ such that correspondingly each subset $A' \in \mathscr{F'}$ has a pre-image  
88
89

$$\begin{equation}
linushof's avatar
linushof committed
90
X^{-1}A' \in \mathscr{F} \; ,
91
92
\end{equation}$$

linushof's avatar
linushof committed
93
which is the set $\{\omega \in \Omega: X(\omega) \in A'\}$ [@kolmogorovFoundationsTheoryProbability1950, p. 21].
94

linushof's avatar
linushof committed
95
B) The mapping is such that $X(\omega) = \omega$. 
96

linushof's avatar
linushof committed
97
98
99
Given conditions A and B, we denote any occurrence of $\omega$ as a *single sample*, or realization, of a random variable defined on a prospect and the process of generating a sequence of single samples as *sampling*. 
Note that, since single samples of a sequence originating from the same prospect are independent and identically distributed (iid), the weak law of the large number applies to the relative frequency of occurrence of an outcome $\omega$ in such a sequence [cf. @bernoulliArsConjectandiOpus1713].
Thus, large sample sequences in principle allow to obtain the same information about a prospect by sampling as by symbolic description.
100

101
## A Stochastical Sampling Model for DfE
102

linushof's avatar
linushof committed
103
104
Consider a choice between prospects $1, ..., k$.
To construct a stochastic sampling model (hereafter SSM) of the random process underlying DfE, we assume that agents base their decision on the information related to these prospects and define a decision variable as a function
105
106

$$\begin{equation}
linushof's avatar
linushof committed
107
108
D:= f((\Omega, \mathscr{F}, P)_1, ..., (\Omega, \mathscr{F}, P)_k)
\;.
109
110
\end{equation}$$

linushof's avatar
linushof committed
111
Since in DfE no symbolic descriptions of the prospects are provided, we restrict the model to the case where decisions are based on sequences of single samples originating from the respective prospects:
112

113
$$\begin{equation}
linushof's avatar
linushof committed
114
D := f(X_{i1}, ..., X_{ik})
115
\; ,
116
117
\end{equation}$$

linushof's avatar
linushof committed
118
where $i = 1, ..., N$ denotes a sequence of length $N$ of random variables that are iid. 
119

linushof's avatar
linushof committed
120
Concerning the form of $f$ and the measures it utilizes, it is quite proper to say that they reflect our assumptions about the exact kind of information agents process and the way they do - not to mention whether they are capable of doing so - and that these choices should be informed by psychological or other theory [@heOntologyDecisionModels2020, for an ontology of decision models] and empirical protocols. 
121
In the following section, it is demonstrated how such assumptions about the processing strategies that agents may apply in DfE can be captured by the SSM.   
122

123
## Integrating sampling and decision strategies into the SSM
124

125
126
127
128
Hills and Hertwig [-@hillsInformationSearchDecisions2010] discussed a potential link between the sampling and decision strategies of agents in DfE, i.e., a systematic relation between the pattern according to which sequences of single samples are generated and the mechanism of integrating and evaluating these sample sequences to arrive at a decision. 
Specifically, the authors suppose that frequent switching between prospects in the sampling phase translates to a round-wise decision strategy, for which the evaluation process is separated into multiple rounds of ordinal comparisons between single samples (or small chunks thereof), such that the unit of the final evaluation are round wins rather than raw outcomes.   
In contrast, infrequent switching is supposed to translate to a decision strategy, for which only a single ordinal comparison of the summaries across all samples of the respective prospects is conducted [@hillsInformationSearchDecisions2010, see Figure 1].
The authors assume that these distinct sampling and decision strategies lead to characteristic patterns in decision behavior and may serve as an additional explanation for the many empirical protocols which indicate that DfE differ from DfD [@wulffMetaanalyticReviewTwo2018, for a meta-analytic review; but see @foxDecisionsExperienceSampling2006]. 
129

130
In the following, choices between two prospects are considered to integrate the assumptions about the sampling and decision strategies from above into the SSM. 
131

132
133
134
Let $X$ and $Y$ be random variables related to the prospects with the probability spaces $(\Omega, \mathscr{F}, P)_X$ and $(\Omega, \mathscr{F}, P)_Y$.
By definition, the decision variable $D$ should quantify the accumulated evidence for one prospect over the other, which Hills and Hertwig [-@hillsInformationSearchDecisions2010] describe in units of won comparisons.
Hence, $f$ should map the possible outcomes of a comparison of quantitative measures related to $X$ and $Y$, hereafter the sampling space $S = \mathbb{R}$, to a measure space $S' = \{0,1\}$, indicating the possible outcomes of a single comparison:
135

136
137
138
139
140
141
$$\begin{equation}
D:= f: S \mapsto S' 
\; .
\end{equation}$$

Since Hills and Hertwig [-@hillsInformationSearchDecisions2010] assume that comparisons of prospects are based on sample means, $S$ is the set
142
143

$$\begin{equation}
144
145
146
147
S = 
  \left\{
    \frac{\frac{1}{N_X} \sum\limits_{i=1}^{N_X} x_i}
    {\frac{1}{N_Y} \sum\limits_{j=1}^{N_Y} y_j} 
148
  \right\} 
149
150
  = 
  \left\{
linushof's avatar
linushof committed
151
    \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}
152
  \right\}
153
  \; ,
154
155
\end{equation}$$

156
where $x_i$ and $y_j$ are single samples and $N_X$ and $N_Y$ denotes the number of single samples within a comparison. 
157
To indicate that the comparison of prospects on the ordinal scale is of primary interest, we define 
158

159
$$\begin{equation}
linushof's avatar
linushof committed
160
\mathscr{D} = \left\{\frac{\overline{X}_{N_X}}{\overline{Y}_{N_Y}} > 0, \frac{\overline{X}_{N_X}}{\overline{Y}_{N_Y}} \leq 0 \right\} 
161
\end{equation}$$
162

163
as a set of subsets of $S$, i.e., the *event space*, and the decision variable as the measure 
164

165
$$\begin{equation}
166
D:= f: (S, \mathscr{D}) \mapsto S'
167
\end{equation}$$
linushof's avatar
linushof committed
168

169
with the mapping
170

171
$$\begin{equation}
172
173
D:= 
  \left(
linushof's avatar
linushof committed
174
    \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}
175
176
177
178
  \right) 
  \in S : 
  f
  \left(
linushof's avatar
linushof committed
179
    \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}
180
181
  \right) 
  =
182
  \begin{cases}
linushof's avatar
linushof committed
183
184
    1 & \text{if} & \frac{\overline{X}_{N_X}}{\overline{Y}_{N_Y}} > 0 \in \mathscr{D} \\
    0 & \text{else}. 
185
  \end{cases} 
186
\end{equation}$$
187

188
189
190
191
192
193
194
195
It can be shown that for the case $N_X = N_Y = 1$, $D$ is a random variable that follows the Bernoulli distribution

$$\begin{equation}
D \sim B\left( p \left( \frac{\overline{X}_{N_X = 1}}{\overline{Y}_{N_Y = 1}} > 0\right), n\right) \; ,
\end{equation}$$

where $n$ is the number of comparisons (see *Proof 1* in Appendix). 

linushof's avatar
linushof committed
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
## Predicting Choices From the SSM

Hills and Hertwig [-@hillsInformationSearchDecisions2010] proposed the two different sampling strategies in combination with the respective decision strategies, i.e., piecewise sampling and round-wise comparison vs. comprehensive sampling and summary comparison, as an explanation for different choice patterns in DfE. 
How does the current version of the SSM support this proposition? 
Given prospects $X$ and $Y$, the sample spaces $S = \left\{\frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}\right\}^{\mathbb{N}}$ can be varied by changes on three parameters, i.e., the number of comparisons $\mathbb{N}$ and the sample sizes $N_X$ and $N_Y$ on which these comparisons are based.
First, only considering the pure cases formulated by the above authors, the following restrictions are put the parameters:

$$\begin{equation}
\mathbb{N}
  =
  \begin{cases}
    1 & \text{if} & \text{Summary} \\ 
    \geq 1 & \text{if} & \text{Round-wise}  
  \end{cases}\\
\end{equation}$$
 
and 

$$\begin{equation}
N_X \, \text{and} \, N_Y
  =
  \begin{cases}
    \geq 1 & \text{if} & \text{Summary} \\ 
    1 & \text{if} & \text{Round-wise}  
  \end{cases}\\
\end{equation}$$

For the summary strategy, the following prediction is obtained: 

Given that 

$$\begin{equation}
228
229
P\left(\lim_{N_X \to \infty} \overline{X}_{N_X} = E(X) \right) = 
P\left(\lim_{N_Y \to \infty} \overline{Y}_{N_Y} = E(Y) \right) = 
linushof's avatar
linushof committed
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
1 \; ,
\end{equation}$$

we obtain that

$$\begin{equation}
\left(
    \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}
\right)
\in S : 
P\left(\lim_{N_X \to \infty} \lim_{N_Y \to \infty} \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}} = 
\frac{E(X)} {E(Y)}
\right ) =
1 \; ,
\end{equation}$$

I.e., for the summary strategy, we assume that for increasing sample sizes $N_X$ and $N_Y$, the prospect with the larger expected value is chosen almost surely.  

For the round-wise strategy, the following prediction is obtained:

250
Given that $N_X$ and $N_Y$ are set to 1, $D$ follows the binomial distribution 
linushof's avatar
linushof committed
251

252
253
254
255
256
257
258
259
260
261
262
263
264
265
$$\begin{equation}
B(k \, | \, p_X, \mathbb{N}) \; ,
\end{equation}$$

where $p$ is the probability that a single sample of prospect $X$ is larger than a single sample of prospect $Y$, $\mathbb{N}$ is the number of comparisons and $k$ is the number of times $x \in X$ is larger than $y \in Y$. 

*Proof.*

For $N_X = N_Y = 1$, the sample space is 

$$\begin{equation}
\left\{\frac{\overline{X}_{N_X = 1}} {\overline{Y}_{N_Y = 1}} \right\}^{\mathbb{N}} =
\left\{\frac{x_i \in \Omega'_X} {y_j \in \Omega'_Y}\right\}^{\mathbb{N}}
\end{equation}$$
266

linushof's avatar
linushof committed
267
268
# Method

linushof's avatar
linushof committed
269
## Test set
270

linushof's avatar
linushof committed
271
272
273
274
275
276
277
278
279
Under each condition, i.e., strategy-parameter combinations, all gambles are played by 100 synthetic agents.
We test a set of gambles, in which one of the prospects contains a safe outcome and the other two risky outcomes (*safe-risky gambles*).
Therefore, 60 gambles from an initial set of 10,000 are sampled.
Both outcomes and probabilities are drawn from uniform distributions, ranging from 0 to 20 for outcomes and from .01 to .99 for probabilities of the lower risky outcomes $p_{low}$.
The probabilities of the higher risky outcomes are $1-p_{low}$, respectively.
To omit dominant prospects, safe outcomes fall between both risky outcomes.
The table below contains the test set of 60 gambles.
Sampling of gambles was stratified, randomly drawing an equal number of 20 gambles with no, an attractive, and an unattractive rare outcome.
Risky outcomes are considered *"rare"* if their probability is $p < .2$ and *"attractive"* (*"unattractive"*) if they are higher (lower) than the safe outcome.
280

linushof's avatar
linushof committed
281
282
283
```{r message=FALSE}
gambles <- read_csv("data/gambles/sr_subset.csv")
gambles %>% kable()
284
285
```

linushof's avatar
linushof committed
286
## Model Parameters
287

linushof's avatar
linushof committed
288
**Switching probability** $s$ is the probability with which agents draw the following single sample from the prospect they did not get their most recent single sample from.
linushof's avatar
linushof committed
289
$s$ is varied between .1 to 1 in increments of .1.
290

linushof's avatar
linushof committed
291
The **boundary type** is either the minimum value any prospect's sample statistic must reach (absolute) or the minimum value for the difference of these statistics (relative).
linushof's avatar
linushof committed
292
Sample statistics are sums over outcomes (comprehensive strategy) and sums over wins (piecewise strategy), respectively.
293

linushof's avatar
linushof committed
294
295
For comprehensive integration, the **boundary value** $a$ is varied between 15 to 75 in increments of 15.
For piecewise integration $a$ is varied between 1 to 5 in increments of 1.
296

linushof's avatar
linushof committed
297
```{r message=FALSE}
298
299
300
301
302
303
304
305
# read choice data 
cols <- list(.default = col_double(),
             strategy = col_factor(),
             boundary = col_factor(),
             gamble = col_factor(),
             rare = col_factor(),
             agent = col_factor(),
             choice = col_factor())
linushof's avatar
linushof committed
306
choices <- read_csv("data/choices/choices.csv", col_types = cols)
307
308
```

linushof's avatar
linushof committed
309
In sum, 2 (strategies) x 60 (gambles) x 100 (agents) x 100 (parameter combinations) = `r nrow(choices)` choices are simulated.
linushof's avatar
linushof committed
310

linushof's avatar
linushof committed
311
# Results
312

linushof's avatar
linushof committed
313
314
Because we are not interested in deviations from normative choice due to sampling artifacts (e.g., ceiling effects produced by low boundaries), we remove trials in which only one prospect was attended.
In addition, we use relative frequencies of sampled outcomes rather than 'a priori' probabilities to compare actual against normative choice behavior.
315
316

```{r}
linushof's avatar
linushof committed
317
318
319
# remove choices where prospects were not attended
choices <- choices %>%
  filter(!(is.na(a_ev_exp) | is.na(b_ev_exp)))
320
321
```

linushof's avatar
linushof committed
322
323
324
325
326
```{r eval = FALSE}
# remove choices where not all outcomes were sampled
choices <- choices %>% 
  filter(!(is.na(a_ev_exp) | is.na(b_ev_exp) | a_p1_exp == 0 | a_p2_exp == 0))
```
linushof's avatar
linushof committed
327

linushof's avatar
linushof committed
328
Removing the respective trials, we are left with `r nrow(choices)` choices.
linushof's avatar
linushof committed
329

linushof's avatar
linushof committed
330
## Sample Size
linushof's avatar
linushof committed
331

linushof's avatar
linushof committed
332
333
334
335
336
337
```{r message=FALSE}
samples <- choices %>% 
  group_by(strategy, s, boundary, a) %>% 
  summarise(n_med = median(n_sample))
samples_piecewise <- samples %>% filter(strategy == "piecewise")
samples_comprehensive <- samples %>% filter(strategy == "comprehensive")
338
339
```

linushof's avatar
linushof committed
340
The median sample sizes generated by different parameter combinations ranged from `r min(samples_piecewise$n_med)` to `r max(samples_piecewise$n_med)` for piecewise integration and `r min(samples_comprehensive$n_med)` to `r max(samples_comprehensive$n_med)` for comprehensive integration.
341

linushof's avatar
linushof committed
342
### Boundary type and boundary value (a)
343

linushof's avatar
linushof committed
344
As evidence is accumulated sequentially, relative boundaries and large boundary values naturally lead to larger sample sizes, irrespective of the integration strategy.
linushof's avatar
linushof committed
345

linushof's avatar
linushof committed
346
347
```{r message=FALSE}
group_med <- samples_piecewise %>%
linushof's avatar
linushof committed
348
  group_by(boundary, a) %>% 
linushof's avatar
linushof committed
349
  summarise(group_med = median(n_med)) # to get the median across all s values
linushof's avatar
linushof committed
350

linushof's avatar
linushof committed
351
352
samples_piecewise %>%
  ggplot(aes(a, n_med, color = a)) + 
linushof's avatar
linushof committed
353
  geom_jitter(alpha = .5, size = 2) +
linushof's avatar
linushof committed
354
355
356
  geom_point(data = group_med, aes(y = group_med), size = 3) +
  facet_wrap(~boundary) + 
  scale_color_viridis() + 
357
  labs(title = "Piecewise Integration",
linushof's avatar
linushof committed
358
       x ="a", 
linushof's avatar
linushof committed
359
       y="Sample Size", 
linushof's avatar
linushof committed
360
       col="a") + 
linushof's avatar
linushof committed
361
  theme_minimal()
linushof's avatar
linushof committed
362
```
linushof's avatar
linushof committed
363

linushof's avatar
linushof committed
364
365
```{r message=FALSE}
group_med <- samples_comprehensive %>%
linushof's avatar
linushof committed
366
  group_by(boundary, a) %>% 
linushof's avatar
linushof committed
367
  summarise(group_med = median(n_med)) 
linushof's avatar
linushof committed
368

linushof's avatar
linushof committed
369
370
samples_comprehensive %>%
  ggplot(aes(a, n_med, color = a)) + 
linushof's avatar
linushof committed
371
  geom_jitter(alpha = .5, size = 2) +
linushof's avatar
linushof committed
372
373
374
  geom_point(data = group_med, aes(y = group_med), size = 3) +
  facet_wrap(~boundary) + 
  scale_color_viridis() + 
375
  labs(title = "Comprehensive Integration",
linushof's avatar
linushof committed
376
       x ="a", 
linushof's avatar
linushof committed
377
       y="Sample Size", 
linushof's avatar
linushof committed
378
       col="a") + 
linushof's avatar
linushof committed
379
  theme_minimal()
380
381
```

linushof's avatar
linushof committed
382
### Switching probability (s)
383

linushof's avatar
linushof committed
384
385
386
For piecewise integration, there is an inverse relationship between switching probability and sample size.
I.e., the lower s, the less frequent prospects are compared and thus, boundaries are only approached with larger sample sizes.
This effect is particularly pronounced for low probabilities such that the increase in sample size accelerates as switching probability decreases.
linushof's avatar
linushof committed
387

linushof's avatar
linushof committed
388
389
```{r message=FALSE}
group_med <- samples_piecewise %>%
linushof's avatar
linushof committed
390
  group_by(boundary, s) %>% 
linushof's avatar
linushof committed
391
  summarise(group_med = median(n_med)) # to get the median across all a values
linushof's avatar
linushof committed
392

linushof's avatar
linushof committed
393
394
395
396
397
398
samples_piecewise %>%
  ggplot(aes(s, n_med, color = s)) + 
  geom_jitter(alpha = .5, size = 2) +
  geom_point(data = group_med, aes(y = group_med), size = 3) +
  facet_wrap(~boundary) + 
  scale_color_viridis() + 
399
  labs(title = "Piecewise Integration",
linushof's avatar
linushof committed
400
       x ="s", 
linushof's avatar
linushof committed
401
       y="Sample Size", 
linushof's avatar
linushof committed
402
       col="s") + 
linushof's avatar
linushof committed
403
404
405
  theme_minimal()
```

linushof's avatar
linushof committed
406
407
408
For comprehensive integration, boundary types differ in the effects of switching probability.
For absolute boundaries, switching probability has no apparent effect on sample size as the distance of a given prospect to its absolute boundary is not changed by switching to (and sampling from) the other prospect.
For relative boundaries, however, samples sizes increase with switching probability.
linushof's avatar
linushof committed
409

linushof's avatar
linushof committed
410
411
```{r message=FALSE}
group_med <- samples_comprehensive %>%
linushof's avatar
linushof committed
412
  group_by(boundary, s) %>% 
linushof's avatar
linushof committed
413
  summarise(group_med = median(n_med)) # to get the median across all a values
linushof's avatar
linushof committed
414

linushof's avatar
linushof committed
415
416
417
418
419
420
samples_comprehensive %>%
  ggplot(aes(s, n_med, color = s)) + 
  geom_jitter(alpha = .5, size = 2) +
  geom_point(data = group_med, aes(y = group_med), size = 3) +
  facet_wrap(~boundary) + 
  scale_color_viridis() + 
421
  labs(title = "Comprehensive Integration",
linushof's avatar
linushof committed
422
423
424
       x ="s",
       y = "Sample Size", 
       col="s") + 
linushof's avatar
linushof committed
425
426
427
  theme_minimal()
```

linushof's avatar
linushof committed
428
## Choice Behavior
linushof's avatar
linushof committed
429

linushof's avatar
linushof committed
430
Below, in extension to Hills and Hertwig [-@hillsInformationSearchDecisions2010], the interplay of integration strategies, gamble features, and model parameters in their effects on choice behavior in general and their contribution to underweighting of rare events in particular is investigated.
linushof's avatar
linushof committed
431
432
433
434
435
436
437
438
439
440
441
442
We apply two definitions of underweighting of rare events: Considering false response rates, we define underweighting such that the rarity of an attractive (unattractive) outcome leads to choose the safe (risky) prospect although the risky (safe) prospect has a higher expected value.

```{r message=FALSE}
fr_rates <- choices %>% 
  mutate(ev_ratio_exp = round(a_ev_exp/b_ev_exp, 2), 
         norm = case_when(ev_ratio_exp > 1 ~ "A", ev_ratio_exp < 1 ~ "B")) %>% 
  filter(!is.na(norm)) %>% # exclude trials with normative indifferent options
  group_by(strategy, s, boundary, a, rare, norm, choice) %>% # group correct and incorrect responses
  summarise(n = n()) %>% # absolute numbers 
  mutate(rate = round(n/sum(n), 2), # response rates 
         type = case_when(norm == "A" & choice == "B" ~ "false safe", norm == "B" & choice == "A" ~ "false risky")) %>% 
  filter(!is.na(type)) # remove correct responses
linushof's avatar
linushof committed
443
444
```

linushof's avatar
linushof committed
445
Considering the parameters of Prelec's [-@prelecProbabilityWeightingFunction1998] implementation of the weighting function [CPT; cf. @tverskyAdvancesProspectTheory1992], underweighting is reflected by decisions weights estimated to be smaller than the corresponding objective probabilities.
linushof's avatar
linushof committed
446

linushof's avatar
linushof committed
447
### False Response Rates
linushof's avatar
linushof committed
448

linushof's avatar
linushof committed
449
450
451
```{r message=FALSE}
fr_rates_piecewise <- fr_rates %>% filter(strategy == "piecewise")
fr_rates_comprehensive <- fr_rates %>% filter(strategy == "comprehensive")
linushof's avatar
linushof committed
452
```
453

linushof's avatar
linushof committed
454
The false response rates generated by different parameter combinations ranged from `r min(fr_rates_piecewise$rate)` to `r max(fr_rates_piecewise$rate)` for piecewise integration and from `r min(fr_rates_comprehensive$rate)` to `r max(fr_rates_comprehensive$rate)` for comprehensive integration.
linushof's avatar
linushof committed
455
However, false response rates vary considerably as a function of rare events, indicating that their presence and attractiveness are large determinants of false response rates.
linushof's avatar
linushof committed
456

linushof's avatar
linushof committed
457
458
459
460
461
462
```{r message=FALSE}
fr_rates %>% 
  group_by(strategy, boundary, rare) %>% 
  summarise(min = min(rate),
            max = max(rate)) %>% 
  kable()
linushof's avatar
linushof committed
463
464
```

linushof's avatar
linushof committed
465
The heatmaps below show the false response rates for all strategy-parameter combinations.
linushof's avatar
linushof committed
466
467
Consistent with our - somewhat rough - definition of underweighting, the rate of false risky responses is generally higher, if the unattractive outcome of the risky prospect is rare (top panel).
Conversely, if the attractive outcome of the risky prospect is rare, the rate of false safe responses is generally higher (bottom panel).
linushof's avatar
linushof committed
468
As indicated by the larger range of false response rates, the effects of rare events are considerably larger for piecewise integration.
469

linushof's avatar
linushof committed
470
471
472
473
474
475
476
477
478
479
480
481
482
483
```{r message=FALSE}
fr_rates %>% 
  filter(strategy == "piecewise", boundary == "absolute") %>% 
  ggplot(aes(a, s, fill = rate)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_tile(colour="white", size=0.25) + 
  scale_x_continuous(expand=c(0,0), breaks = seq(1, 5, 1)) +
  scale_y_continuous(expand=c(0,0), breaks = seq(.1, 1, .1)) +
  scale_fill_viridis() + 
  labs(title = "Piecewise Integration | Absolute Boundary",
       x = "a", 
       y= "s", 
       fill = "% False Responses") + 
  theme_minimal() 
484
485
```

linushof's avatar
linushof committed
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
```{r message=FALSE}
fr_rates %>% 
  filter(strategy == "piecewise", boundary == "relative") %>% 
  ggplot(aes(a, s, fill = rate)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_tile(colour="white", size=0.25) + 
  scale_x_continuous(expand=c(0,0), breaks = seq(1, 5, 1)) +
  scale_y_continuous(expand=c(0,0), breaks = seq(.1, 1, .1)) +
  scale_fill_viridis() + 
  labs(title = "Piecewise Integration | Relative Boundary",
       x = "a", 
       y= "s", 
       fill = "% False Responses") + 
  theme_minimal() 
```
linushof's avatar
linushof committed
501

linushof's avatar
linushof committed
502
503
```{r message=FALSE}
fr_rates %>% 
linushof's avatar
linushof committed
504
  filter(strategy == "comprehensive", boundary == "absolute") %>% 
linushof's avatar
linushof committed
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
  ggplot(aes(a, s, fill = rate)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_tile(colour="white", size=0.25) + 
  scale_x_continuous(expand=c(0,0), breaks = seq(15, 75, 15)) +
  scale_y_continuous(expand=c(0,0), breaks = seq(.1, 1, .1)) +
  scale_fill_viridis() + 
  labs(title = "Comprehensive Integration | Absolute Boundary",
       x = "a", 
       y= "s", 
       fill = "% False Responses") + 
  theme_minimal() 
```

```{r message=FALSE}
fr_rates %>% 
linushof's avatar
linushof committed
520
  filter(strategy == "comprehensive", boundary == "relative") %>% 
linushof's avatar
linushof committed
521
522
523
524
525
526
527
528
529
530
531
  ggplot(aes(a, s, fill = rate)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_tile(colour="white", size=0.25) + 
  scale_x_continuous(expand=c(0,0), breaks = seq(15, 75, 15)) +
  scale_y_continuous(expand=c(0,0), breaks = seq(.1, 1, .1)) +
  scale_fill_viridis() + 
  labs(title = "Comprehensive Integration | Relative Boundary",
       x = "a", 
       y= "s", 
       fill = "% False Responses") + 
  theme_minimal() 
532
533
```

linushof's avatar
linushof committed
534
#### Switching Probability (s) and Boundary Value (a)
linushof's avatar
linushof committed
535

linushof's avatar
linushof committed
536
As for both piecewise and comprehensive integration the differences between boundary types are rather minor and of magnitude than of qualitative pattern, the remaining analyses of false response rates are summarized across absolute and relative boundaries.
linushof's avatar
linushof committed
537

linushof's avatar
linushof committed
538
Below, the $s$ and $a$ parameter are considered as additional sources of variation in the false response pattern above and beyond the interplay of integration strategies and the rarity and attractiveness of outcomes.
linushof's avatar
linushof committed
539

linushof's avatar
linushof committed
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
```{r message=FALSE}
fr_rates %>% 
  filter(strategy == "piecewise") %>% 
  ggplot(aes(s, rate, color = a)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_jitter(size = 2) + 
  scale_x_continuous(breaks = seq(0, 1, .1)) +
  scale_y_continuous(breaks = seq(0, 1, .1)) +
  scale_color_viridis() + 
  labs(title = "Piecewise Integration",
       x = "s", 
       y= "% False Responses", 
       color = "a") + 
  theme_minimal() 
```
555

linushof's avatar
linushof committed
556
557
```{r message=FALSE}
fr_rates %>% 
linushof's avatar
linushof committed
558
  filter(strategy == "comprehensive") %>% 
linushof's avatar
linushof committed
559
560
561
562
563
564
  ggplot(aes(s, rate, color = a)) + 
  facet_grid(type ~ fct_relevel(rare, "attractive", "none", "unattractive"), switch = "y") +
  geom_jitter(size = 2) + 
  scale_x_continuous(breaks = seq(0, 1, .1)) +
  scale_y_continuous(breaks = seq(0, 1, .1)) +
  scale_color_viridis() + 
565
  labs(title = "Comprehensive Integration",
linushof's avatar
linushof committed
566
567
568
569
       x = "s", 
       y= "% False Responses", 
       color = "a") + 
  theme_minimal() 
570
571
```

linushof's avatar
linushof committed
572
For piecewise integration, switching probability is naturally related to the size of the samples on which the round-wise comparisons of prospects are based on, with low values of $s$ indicating large samples and vice versa.
linushof's avatar
linushof committed
573
Accordingly, switching probability is positively related to false response rates.
linushof's avatar
linushof committed
574
575
I.e., the larger the switching probability, the smaller the round-wise sample size and the probability of experiencing a rare event within a given round.
Because round-wise comparisons are independent of each other and binomial distributions within a given round are skewed for small samples and outcome probabilities [@kolmogorovFoundationsTheoryProbability1950], increasing boundary values do not reverse but rather amplify this relation.
576

linushof's avatar
linushof committed
577
578
579
For comprehensive integration, switching probability is negatively related to false response rates, i.e., an increase in $s$ is associated with decreasing false response rates.
This relation, however, may be the result of an artificial interaction between the $s$ and $a$ parameter.
Precisely, in the current algorithmic implementation of sampling with a comprehensive integration mechanism, decreasing switching probabilities cause comparisons of prospects based on increasingly unequal sample sizes immediately after switching prospects.
linushof's avatar
linushof committed
580
Consequentially, reaching (low) boundaries is rather a function of switching probability and associated sample sizes than of actual evidence for a given prospect over the other.
581

linushof's avatar
linushof committed
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
### Cumulative Prospect Theory

In the following, we examine the possible relations between the parameters of the *choice-generating* sampling models and the *choice-describing* cumulative prospect theory.

For each distinct strategy-parameter combination, we ran 20 chains of 40,000 iterations each, after a warm-up period of 1000 samples.
To reduce potential autocorrelation during the sampling process, we only kept every 20th sample (thinning).

```{r}
# read CPT data
cols <- list(.default = col_double(),
             strategy = col_factor(),
             boundary = col_factor(),
             parameter = col_factor())
estimates <- read_csv("data/estimates/estimates_cpt_pooled.csv", col_types = cols)
```

#### Convergence

```{r}
gel_92 <- max(estimates$Rhat) # get largest scale reduction factor (Gelman & Rubin, 1992) 
```

The potential scale reduction factor $\hat{R}$ was $n \leq$ `r round(gel_92, 3)` for all estimates, indicating good convergence.

606
#### Piecewise Integration
linushof's avatar
linushof committed
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640

```{r}
# generate subset of all strategy-parameter combinations (rows) and their parameters (columns)
curves_cpt <- estimates %>% 
  select(strategy, s, boundary, a, parameter, mean) %>% 
  pivot_wider(names_from = parameter, values_from = mean)
```

##### Weighting function w(p)

We start by plotting the weighting curves for all parameter combinations under piecewise integration.

```{r}

cpt_curves_piecewise <- curves_cpt %>% 
  filter(strategy == "piecewise") %>% 
  expand_grid(p = seq(0, 1, .1)) %>% # add vector of objective probabilities
  mutate(w = round(exp(-delta*(-log(p))^gamma), 2)) # compute decision weights (cf. Prelec, 1998)

# all strategy-parameter combinations 

cpt_curves_piecewise %>% 
  ggplot(aes(p, w)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Piecewise Integration: Weighting functions",
       x = "p", 
       y= "w(p)") + 
  theme_minimal() 
```

```{r}
cpt_curves_piecewise %>% 
  ggplot(aes(p, w)) + 
641
  geom_path() +
linushof's avatar
linushof committed
642
643
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
644
645
646
647
648
  labs(title = "Piecewise Integration: Weighting functions",
       x = "p",
       y= "w(p)",
       color = "Switching Probability") + 
  scale_color_viridis() +
linushof's avatar
linushof committed
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
  theme_minimal() 
```

```{r}
cpt_curves_piecewise %>% 
  ggplot(aes(p, w, color = s)) + 
  geom_path() +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Piecewise Integration: Weighting functions",
       x = "p", 
       y= "w(p)", 
       color = "Switching Probability") + 
  scale_color_viridis() +
  theme_minimal() 
```

```{r}
cpt_curves_piecewise %>% 
  ggplot(aes(p, w, color = s)) + 
  geom_path() +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
  labs(title = "Piecewise Integration: Weighting functions",
       x = "p",
       y= "w(p)",
       color = "Switching Probability") + 
  scale_color_viridis() +
  theme_minimal() 
```

##### Value function v(x)

```{r}

cpt_curves_piecewise <- curves_cpt %>% 
  filter(strategy == "piecewise") %>% 
  expand_grid(x = seq(0, 20, 2)) %>% # add vector of objective outcomes
  mutate(v = round(x^alpha, 2)) # compute decision weights (cf. Prelec, 1998)

# all strategy-parameter combinations 

cpt_curves_piecewise %>% 
  ggplot(aes(x, v)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Piecewise Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  theme_minimal() 
```

```{r}
cpt_curves_piecewise %>% 
  ggplot(aes(x, v, color = s)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Piecewise Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  scale_color_viridis() + 
  theme_minimal() 
```

```{r}
cpt_curves_piecewise %>% 
  ggplot(aes(x, v, color = s)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
  labs(title = "Piecewise Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  scale_color_viridis() + 
  theme_minimal() 
```

725
#### Comprehensive Integration
linushof's avatar
linushof committed
726
727
728
729

##### Weighting function w(p)

We start by plotting the weighting curves for all parameter combinations under piecewise integration.
linushof's avatar
linushof committed
730

linushof's avatar
linushof committed
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
```{r}

cpt_curves_comprehensive <- curves_cpt %>% 
  filter(strategy == "comprehensive") %>% 
  expand_grid(p = seq(0, 1, .1)) %>% # add vector of objective probabilities
  mutate(w = round(exp(-delta*(-log(p))^gamma), 2)) # compute decision weights (cf. Prelec, 1998)

# all strategy-parameter combinations 

cpt_curves_comprehensive %>% 
  ggplot(aes(p, w)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Comprehensive Integration: Weighting functions",
       x = "p", 
       y= "w(p)") + 
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
  ggplot(aes(p, w)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Comprehensive Integration: Weighting functions",
       x = "p", 
757
758
       y= "w(p)") + 
  facet_wrap(~a) + 
linushof's avatar
linushof committed
759
760
761
762
763
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
764
765
  ggplot(aes(p, w, color = s)) + 
  geom_path() +
linushof's avatar
linushof committed
766
767
768
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Comprehensive Integration: Weighting functions",
       x = "p", 
769
770
771
       y= "w(p)", 
       color = "Switching Probability") + 
  scale_color_viridis() +
linushof's avatar
linushof committed
772
773
774
775
776
777
778
779
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
  ggplot(aes(p, w, color = s)) + 
  geom_path() +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
780
  facet_wrap(~a) + 
linushof's avatar
linushof committed
781
  labs(title = "Comprehensive Integration: Weighting functions",
782
783
       x = "p",
       y= "w(p)",
linushof's avatar
linushof committed
784
785
786
787
788
789
790
       color = "Switching Probability") + 
  scale_color_viridis() +
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
791
  filter(s >= .7) %>% 
linushof's avatar
linushof committed
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
  ggplot(aes(p, w, color = s)) + 
  geom_path() +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
  labs(title = "Comprehensive Integration: Weighting functions",
       x = "p",
       y= "w(p)",
       color = "Switching Probability") + 
  scale_color_viridis() +
  theme_minimal() 
```

##### Value function v(x)

```{r}

cpt_curves_comprehensive <- curves_cpt %>% 
  filter(strategy == "comprehensive") %>% 
  expand_grid(x = seq(0, 20, 2)) %>% # add vector of objective outcomes
  mutate(v = round(x^alpha, 2)) # compute decision weights (cf. Prelec, 1998)


# all strategy-parameter combinations 

cpt_curves_comprehensive %>% 
  ggplot(aes(x, v)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Comprehensive Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
  ggplot(aes(x, v)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
  labs(title = "Comprehensive Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
  ggplot(aes(x, v, color = s)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  labs(title = "Comprehensive Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  scale_color_viridis() + 
  theme_minimal() 
```

```{r}
cpt_curves_comprehensive %>% 
  ggplot(aes(x, v, color = s)) + 
  geom_path(size = .5) +
  geom_abline(intercept = 0, slope = 1, color = "red", size = 1) +
  facet_wrap(~a) + 
  labs(title = "Comprehensive Integration: Value functions",
       x = "p", 
       y= "w(p)") + 
  scale_color_viridis() + 
  theme_minimal() 
```
linushof's avatar
linushof committed
862

863
864
865
866
# Discussion 

# Conclusion

867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
# Appendix 

Let $X_n$ and $Y_m$ be independent and discrete random variables of the sequences

$$\begin{equation}
X_1, ..., X_n, ..., X_{N_X}
\end{equation}$$

and 

$$\begin{equation}
Y_1, ..., Y_m, ..., Y_{N_Y} \; .
\end{equation}$$

Then

$$\begin{equation}
P\left(\frac{\overline{X}_{N_X}}{\overline{Y}_{N_Y}} > 0 \right) = 
P\left(\frac{\frac{1}{N_X}\sum\limits_{n=1}^{N_X} (X(\omega_i) = A'_X \in \mathscr{F'}_X)_n} 
{\frac{1}{N_Y}\sum\limits_{m=1}^{N_Y} (Y(\omega_j) = A'_Y \in \mathscr{F'}_Y)_m} > 0 \right)
\end{equation}$$

is the probability that the quotient of the mean of both sequences takes on a value larger $0$.
Given the sample sizes $N_X = N_Y = 1$, the equation reduces to 

$$\begin{equation}
P\left(\frac{X(\omega_i \in \Omega_X) = A'_X \in \mathscr{F'}_X}{Y(\omega_j \in \Omega_Y) = A'_Y \in \mathscr{F'}_Y} > 0 \right) \; ,
\end{equation}$$

which is the sum across all joint probabilities $p(\omega_i \cap \omega_j)$ for which the above inequation holds:.


$$\begin{equation}
D:= 
  f
  \left(
    \frac{\overline{X}_{N_X}} {\overline{Y}_{N_Y}}
  \right) 
  =
  \begin{cases}
    1 & \text{if} & \frac{\overline{X}_{N_X}}{\overline{Y}_{N_Y}} > 0 \in \mathscr{D} \\
    0 & \text{else}. 
  \end{cases} 
\end{equation}$$

linushof's avatar
linushof committed
912
# References
913