The black box method was developed as an ‘asocial control’ to measure the potential role of payoff-based learning in social dilemmas [1]. Players must decide how many virtual coins they want to input into a virtual black box that will provide uncertain returns. But in truth, they are playing with each other in a repeated social game. By ‘black boxing’ the game’s social aspects and payoffs, the method creates a population of self-interested but ignorant or confused individuals that must learn the games payoffs. This provides a behaviourally measured null hypothesis for testing social behaviours, as opposed to the idealized and stringent predictions of rational self-interested agents (Homo economicus). However, a potential problem is that participants can unwittingly affect other participants. Here we test a solution to this problem, in a range of public goods games, by making participants interact, unknowingly, with simulated players (‘computerized black box’). We find no significant differences in rates of learning between the original and the computerized black box. These results, along with the fact that simulated agents can be programmed to behave in different ways, mean that the computerized black box has great potential for studying how individuals learn under different environments in social dilemmas.