As just discussed, your base payment for each game will be the average amount your rule and your counterpart's rule would earn over millions of games of varying lengths. However, we will also have you and your counterpart implement the rules you each chose in one actual multi-round game.
Example:
In each game, after you and your counterpart have both chosen a rule, we will show you a screen like the one above. It works exactly as in the games from Part 1 of the experiment. The only difference is that instead of playing with a robot, you will be playing with your actual counterpart. You will be implementing your rule (in response to his or her actions) and your counterpart will be implementing his or her rule (in response to your actions)!
On your screen we will remind you of what rule you chose for the game (we will not show you the rule your counterpart chose). Just as in Part 1 of the experiment, you will lose 20 points every time you make a mistake in implementing your rule.
Remember: although your and your counterpart's rule choices influence your base payment via the actions they generate and the numbers in the game's payment table (like the one below), once you move to implementing the rule, your job is only to try to follow the rule you chose. After choosing your rule, the only impact your actions have on your payments for the rest of the game is to produce possible penalties to your earnings if you make mistakes in following the rule.