Hertwig, R., Woike, J. K., Pachur, T., & Brandstätter, E. (2019). The robust beauty of heuristics in choice under uncertainty. In R. Hertwig, T. Pleskac, T. Pachur, & the Center for Adaptive Rationality (Eds.), Taming uncertainty (pp. xx–xx). Boston, MA: MIT Press. doi:XXXXXXX

## Introduction

This interactive element allows you to experience the simulation described in Chapter 2. There are three different ways to run a simulation:

1. Description/Risk simulates decisions with known probabilities.
2. Experience: Sampling simulates a sequence of outcomes that the strategies can use to estimate probabilities.
3. Experience: Uncertainty allows you to replicate the chapter’s simulation for a set of options.

All three simulations choose between four options with two outcomes shown below. You can change the probabilities with which each of the two outcomes occurs (p). If you decrease the probability of one of the outcomes, the other will increase. In addition, you can determine the values of each of the two outcomes. To change all probabilities and values to those of one of the four prepared examples, click on an example preset—try all four. The strategies featured in the simulations correspond to the strategies described in the chapter.

The current mode allows you to simulate decisions under risk. Five strategies have access to all values and probabilities. You will see results for the four options when you enter a value. The numbers in each row tell you, how many times out of 100 each of the strategies would choose option A, B, C and D. If there is one best option, this number will be 100 for this option and 0 for the other options. If the strategy decides that two or more options are equivalent, the strategy will make a random choice, and 100 will be divided by the number of best options. The bottom row shows the expected value for each of the four options (this value is independent of the strategies’ decisions). The right column shows the average expected values for the decisions that the strategies make (if a strategy is indifferent between more than one options, this will be the average of the expected values of all of these best options).

p
Option A
Option B
Option C
Option D
Value
Strategy Option A Option B Option C Option D Average
expected value
Maximum expected value
Equiprobable
Least likely
Lexicographic
Probable
Expected Value
N = 0

Observation Option A Option B Option C Option D
Observed value
Reconstruction Option A Option B Option C Option D
p1
p2
Strategy Option A Option B Option C Option D
Natural mean
Equiprobable
Least likely
Lexicographic
Probable
N = 0
Strategy Option A Option B Option C Option D Average
expected value
Natural mean
Equiprobable
Least likely
Lexicographic
Probable
Expected value
×

In this element, you gained firsthand experience of how different decision strategies perform when they must reach a decision. We show in the chapter that under uncertainty some heuristics—surprisingly, those that ignore probabilities—perform almost as well (and sometimes even better) than those that use a maximization calculus.

You also experienced how simulation can be a wonderful tool to explore complex problems and to test and challenge intuitions about them. By varying elements of simulations, one can gain insight into the conditions (e.g., environments, sample sizes, problem types) under which observed relationships are observable and thus learn about the robustness or fragility of results. Often simulations can supplement both mathematical analyses and empirical studies. You have earned a badge! Click me to read more.

Click on any reward badge in the footer to review take home messages of this or other elements.