Welcome! Thanks for participating. The purpose of this session is to study how people make decisions. The session will last about 90 minutes. All payments in this experiment will be expressed in points. In a later announcement on the Virtual Campus, we will publish how these points translate into marks for the subject. As the number of points you will get depends on your decisions, it is very important that you read the instructions carefully. After you finish reading these instructions, we will ask you some very simple control questions. If you do not answer the control questions correctly, you may not be able to participate in the experiment and you may not get any points.
You have {{ timeout_mins }} minutes to read the instructions and answer the control questions below. There is more than enough time. You can check the remaining time at the top of this page.
In this session you are going to play {{ player.subsession.num_games }} games, one after another. In each game, the central server will randomly and anonymously divide all the participants into small groups of 3 or 4 people (depending on the game), who will play the game among themselves . For example, if for the first game, the system divided the participants into groups of 3, this game will be played in each group between the 3 people that form the group. Furthermore, since it is a random assignment, no participant will know who the other participants in their group are. And this information will not be revealed throughout the experiment. In each group, each participant will be randomly given the role of player A, B, C or D.
Each of the players in each group has an ideal value that he/she must guess and declare to the system. All players in each group will receive the same information on how the ideal values of all players in their same group are determined. Based on this information, each participant must declare a number (which we call hid/her decision). Declared numbers must be between 0 and 100, with up to two decimal places. You will not know the decisions of the other members of your group when you declare your number, just as no one in your group will know your number when they declare theirs. Keep in mind that the ideal value you have to guess will depend on the numbers declared by the other participants in your group and, therefore, you will not know your ideal value when you make your decision: your goal is to guess it, that is , declare a number that is as close as possible to your ideal value.
In each game, your score is easy to calculate. You will start with 100 points in each game and 1 point will be subtracted for each unit of distance from your decision to your ideal value. That is, if your decision in a particular game was X and your ideal value is V, then your score in that game will be 100-d(X,V), where d(X,V) is the distance between the numbers X and V. Therefore, your decision X must be as close as possible to your ideal value V (which you do not know since you depend on the decisions of others).
Now we are going to provide an example for you to better understand the game.
Please note that the games you will be playing during this experiment are different from the example below. We only present this example as an illustration so that you understand how the game works.
We present an example where groups of three people have been created. Therefore, in each group a game of three players A, B and C is developed. The description of the ideal values given to the players is as follows:
Please note that this full description is provided to all 3 participants in this group: each player receives the full description of how the ideal values of all players in his group are calculated. Therefore, in each game, all participants receive exactly the same information.
{% if player.session.config.show_networks %}The diagram describes this situation. It shows the players in the group (A, B and C in this example), as well as the numbers used in the calculation of the ideal values (the number 20 for this example). The arrows summarize how the ideal values depend on the numbers declared by the other members of the group and on the number 20. For example, the ideal value of A depends on what B declares. Therefore, there is an arrow going from A to B. Similarly, there are two arrows coming out from C and going towards A and 20, respectively, since the ideal value of C depends on what A declares and on 20. Note that the diagram gives information on the dependencies , but not of the specific operations that have to be carried out to calculate the ideal values. These operations appear in the description of the ideal values, which will always be available and can be consulted at any time.
{% endif %}Let us imagine that -continuing with the same example- the decisions made by the three players have been the following:
Therefore, A has declared 40 because he/she estimates that his ideal value is 40. The same applies to players B and C, with their respective declarations 100 and 0. Furthermore, as we have already indicated, no player in the group knows the decisions of the other two players. Otherwise, it would be very easy for you to guess the ideal value.
Once the game is over, the system determines the score of the 3 players in the group, first calculating the ideal value of each player as follows:
Once the ideal values have been calculated, the system calculates the distance between what was declared by each player and his/her ideal value:
Finally, the system calculates the score for each player in this game, which is 100 minus the distance to the ideal value. That is to say ,
Imaginemos que -siguiendo con el mismo ejemplo- las decisiones tomadas por los tres jugadores han sido las siguientes:
Therefore, your objective is to declare a number that is as close as possible to your ideal value, which you do not know as it depends on what the other participants in your group declare.
The structure of the experiment is very simple. The experiment is divided into {{ player.subsession.num_games }} games, each of which is divided into {{ player.session.config.num_rounds_per_game }} periods. Each game will have a different background color, so that you are aware when you've advanced to the next game.
Each of the {{ player.subsession.num_games }} games you will play is divided into {{ player.session.config.num_rounds_per_game}} periods. Your group will be the same during the periods covered by each game.
The duration of each period will be randomly chosen by the central server. No participant will know the exact duration of any of the periods, that is, no one will know when each period ends and each period will end without notice. Due to space limitations, in each game you can declare up to a maximum of {{ player.session.config.max_submissions_per_game}} decisions, which you can register in the system as your estimation of your ideal value changes throughout the game. However, the system will only consider your last decision in each period as valid. The reason for using this mechanism is that we need you to register your decision whenever it changes because, for example, you have performed new calculations. For example, if in a certain period you had entered the number 30 and -after thinking it over- you think that 35 is a more accurate estimate of your ideal value, then you should enter 35 before the period suddenly ends. In this way you ensure that when the period ends (you don't know when) the number registered in the system is 35 and not 30. In short: if during the current period you come up with a number that you think is closer to your ideal value, it is convenient that you declare it, since the system will only take into account the last number that you have entered in each period. This does not mean that you have to register numbers indiscriminately until you reach the maximum of {{ player.session.config.max_submissions_per_game}} decisions per game: doing that would not help you at all since only the last number of each period is taken into account to calculate your score. And it also doesn't mean that after each period the game is restarted: the game continues normally after the end of each period (except for the last period of the game) and you can safely continue with your calculations. To sum up, the best advice we can give you is to not worry about periods and to focus more on thinking about the game you have on your screen. Make your decisions calmly, registering a new decision when you think it is better than the previous one. That's all!
What happens if you don't enter any number in some period? It won't affect you at all. In this case, the system will consider that your decision has not changed from the previous period of the same game, and therefore, it will take your last decision in the previous period as the decision for the current period. There is an exception: if you have not made any decision at the end of first period, then -since there is no period before the first- the system will choose a random number for you. In short, if you don't enter any numbers, nothing happens, the system will use your decision from the previous period of the same game. The only thing that matters to us as researchers is that you enter numbers as your estimate of your ideal value changes. Therefore, do your calculations calmly and update your decision when you think it is necessary. Even if the period ends, the last number you declared is passed to the next period, and the game continues. There is plenty of time to think about each game.
In this experiment, neither you nor the other members of your group will receive any information about the decisions of the others, neither about your score nor that of the others. After the final exam of the subject, we will provide you with the score obtained.
In order to determine your score in each game, at the end of the experiment the central server will randomly select one of the periods of that game, and your score for that game will be the result you obtained in the period selected by the server. Therefore, it is advisable that you play the best you can in all periods. Finally, your final score obtained in the experiment will be the sum of the scores obtained in each of the games.
At the end of the experiment we will give you a short test that you have to complete in order to redeem your points.