Appendix A
Instructions used in the experiment
This appendix contains the written instructions that were distributed among participants of the experiment described in Chapter 4, translated into English. The experiment was conducted in Dutch; the original Dutch instructions are available from the author upon request. In the experiment we used four different versions of the written instructions—two for the two payoff conditions (high risk and low risk) and two for the different information conditions. We present the instructions for the high risk, local information condition. The instructions for the low-risk condition differ only from those in the high-risk condition with regard to the payoffs shown in Figure A.1. In the instructions for the global information condition some sentences were formulated somewhat differently as compared with the local information condition. In the text below, these differences are indicated by footnotes. The numbering of tables and figures in the instructions has been adapted from the original version to be in line with the rest of the dissertation; otherwise, the layout has been preserved as much as possible.
—INSTRUCTIONS START HERE—
Figure A.1 Choices and points.
Instruction
Part 1
Welcome to this experiment, and thank you for coming. Please read the instructions until the bottom of page 7, marked “End of Instruction Part 1.” If you at that point have any questions, feel free to ask for additional assistance by raising your hand. Also during the experiment, you can always ask for assistance by raising your hand in case something is not clear to you. However, please only do so if absolutely necessary; the instruction should speak for itself. All other participants receive the same instructions.
In this experiment you can earn money depending on the choices you make. These earnings will be paid at the end of the experiment, without the other participants knowing how much you earned or which choices you made. From now on, you are no longer allowed to speak with anyone this room, and you should not look at the computer screens of the other participants. It is important that we know your choices based on your own considerations, and that you make your choices based only on this instruction and on what you see at your screen. There are no “right” or “wrong” choices.
During the experiment you will be part of a group with seven other participants. Some of these participants will be your “neighbors,” others will not (these are not necessarily the participants sitting next to you; we get back to this issue later). The experiment consists of two parts. Each part consists of 15 rounds. In each round, you and the other participants in your group make choices. The number of points that you earn depends on both your choices and the choices of your neighbors in your group. For each point, you will be paid € 0.01. We now explain the choice alternatives in detail.
Choices
In each round, you and the others in your group choose simultaneously between “Left” or “Right.” If you choose Left, you earn 20 points for each neighbor who also chooses Left; you earn 0 points for every neighbor who chooses Right. If you choose Right, you earn 16 points for every neighbor in the group who also chooses Right; you earn 14 points for every neighbor who chooses Left. The choices of participants who are not neighbors do not affect your score.
The number of points that you earn with each of your neighbors is summarized in Figure A.1.
We will clarify this later using an example.
Relations and neighbors
Who of the other participants in your group are your neighbors is decided by you. Per round, you can change one relation by proposing a new relation or by ending the relation with a current neighbor. You can become another participant's neighbor only if this participant also wants to become your neighbor; to end a current relation you do not need the permission of the other. You can become neighbors with another participant if one of you proposes this and the other accepts the proposal or if both of you propose simultaneously. You can lose a neighbor if this neighbor does no longer want to be your neighbor, regardless of whether you wanted to remain a neighbor of this participant yourself. It will be explained below how this works in practice.
Lastly, you can not only earn points in relations, but you also have to pay “maintenance costs” for each of your relations. The more relations you have, the more expansive these will be. The costs that you have to pay for each number of neighbors are shown in Table A.1.
Table A.1 Costs of relations
| Number of neighbors | Total costs of your relations |
| 1 | 7 |
| 2 | 16 |
| 3 | 27 |
| 4 | 40 |
| 5 | 55 |
| 6 | 72 |
| 7 | 91 |
Thus, the total profit in one round depends on
- The number of points as depending the choices you and your neighbors make;
- The total costs of your relations.
- You choose Left;
- Neighbor 1 chooses Left;
- Neighbor 2 chooses Right;
- Neighbor 3 chooses Left;
Note: because in every round, you do not only earn points, but also have to pay costs, it is possible that you make a loss in some rounds.
Making choices using the computer
Shortly, you will be matched to seven other participants by the computer. You make your choices using the computer screen. The experiment consists of 15 rounds. Each round consists of the following steps.
The only exception to this is the first round, in which the computer determines who your neighbors are.
In each phase, it is displayed on screen who your neighbors are and what the choices of all other participants in the previous round were. This might, for example, look as in Figure A.2.
Figure A.2 Screenshot of the experimental interface: choosing relations with local information (translated from the original Dutch).
This is the screen in which you can change relations (Step 3). The first column shows the numbers identifying the other subjects, with yourself (number 1) at the top. Note: the numbers 2, 3, etc. refer to the second and third neighbor in your table. Number 2 in your table is not necessarily the same participant as number 2 in the table of one of the other participants. All participants really are people in this room.
The second column displays the behavioral choices of your neighbors (and yourself) in the previous round.1 At moments at which you can change relations (as in the example screen above), this is displayed in the fourth column. To make it a bit easier to see who your neighbors are, all information about your neighbors is printed in bold face.
At the bottom of the screen, we summarize what happened in previous rounds. For each other participant, it is shown whether this participant was a neighbor and if so, what his or her behavior was.2 Your own behavioral choices and payoffs are displayed as well.
Before we start with the actual experiments, we will play a number of practice rounds. In these practice rounds, you will not win or lose any points. Also, you will not be matched in a group with other participants but instead you will be playing against the computer. The computer's choices were pre-programmed. After the practice rounds, we will start with Part 1 of the experiment.
This concludes the instructions for Part 1; you will read the instructions for Part 2 only after Part 1 of the experiment has finished. If you have any questions at this point, please raise your hand now. If all is clear to you, please click “Continue” on the computer screen. As soon as all participants have finished reading the instructions, we will start the practice rounds.
On the next page, we again show Figure 1.1 and Table 1.1. During the experiment, you may use this page for reference.
Figure A.1 Choices and Points.
Table A.1 Costs of relations
| Number of neighbors | Total costs of your relations |
| 1 | 7 |
| 2 | 16 |
| 3 | 27 |
| 4 | 40 |
| 5 | 55 |
| 6 | 72 |
| 7 | 91 |
— END OF THE INSTRUCTIONS FOR PART 1 —
— PART 2 —
Continue reading only if you have been instructed on the screen to do so!
Instruction Part 2
In Part 2, you will again be placed in a group with seven other people in this room; in part, this will be different people than in Part 1. Otherwise, Part 2 will proceed as Part 1. However, there is one difference. In Part 1, you were shown the behavior of your neighbors in the previous round. In Part 2, you will be able to see the behavior of all other participants in the previous round.3 Otherwise, Part 2 is identical to Part 1; you may again use the scheme on page 7 for the scoring of points and the costs of relations. Just like in Part 1, the computer will determine who your neighbors will be in the first round.
When you have finished reading, please click “Continue.”
—END OF THE INSTRUCTIONS—
1. In the global information condition, this sentence read “The second column displays the behavioral choices of each participant (and yourself) in the previous round.”
2. In the instructions for the global information condition, this sentence read: “For each other participant, it is shown whether this participant was a neighbor and what his or her behavior was.”
3. In the instructions for the global information condition, this line read: “In Part 2, you will be able to see only your neighbors’ behavior in the previous round.”