Design and Procedures. Since we wish to compare different measures of pro-social behavior, our specific goals are to determine how decisions in the binary trust game compare to decisions in the investment game and how each of these choices compares to decisions in the public goods game, which measures cooperation. We designed an experiment where the same set of subjects participated in a series of games and surveys. Among the games, we included the binary trust game, the investment game, and the public goods game. The surveys were designed to collect demographic and association/membership information about our subjects. Figure 3 shows the binary trust game, which provides the most basic experimental measure of trust. Here, the first player, Player 1, has the task to either stop (S) or continue (C) the game. If Player 1 stops, then the game is over, and each player receives an amount equal to s. However, if she continues, then Player 2 can choose to either reciprocate (R) or exploit (E). The sub-game perfect equilibrium in this game is for Player 2 to choose E and for Player 1 to choose S, because payoff values are r<s<t<u. In our experiment, r = 5, s = 10, t = 15, and u = 20. The payoffs were chosen so as to match the incentives in the investment game described below. A decision by Player 1 to continue the game represents confidence that Player 2 will reciprocate towards a mutually beneficial outcome. Similarly, a Player 2’s decision to choose R represents trustworthiness. [Insert Figure 3 about here.] While the binary trust game does provide insights into the presence or absence of trust and trustworthiness, its binary nature prohibits a measure of the degree of trust or trustworthiness one player has towards another. In contrast, the investment game is a dynamic game of complete information between two players with identical endowments, say x . The first player, called the investor, faces a choice to transfer to the second player, or trustee, any amount x ∈[0, x] . The experimenter multiplies x by t>1, and the trustee has an option to transfer back any amount y ∈[0, tx] . The payoffs for the first and second players are x − x + y and x + tx − y , respectively.29 While game theoretic assumptions suggest no transfers should occur in equilibrium, in the lab subjects frequently transfer positive sums of money. Commonly the amount, x, transferred by the first player represents her level of trust, and the percentage returned, y/tx, signifies her degree of trustworthiness. In our exp...
