This paper investigates whether there is a connection between psychopathy and certain manifestations of social and economic behavior, measured in a lab-in-the-field experiment with prison inmates. In order to test this main hypothesis, we let inmates play four games that have often been used to measure prosocial and antisocial behavior in previous experimental economics literature. Specifically, they play a prisoner's dilemma, a trust game, the equality equivalence test that elicits distributional preferences, and a corruption game. Psychopathy is measured by means of the Levenson Self-Report Psychopathy Scale (LSRP) questionnaire, which inmates filled out after having made their decisions in the four games. We find that higher scores in the LSRP are significantly correlated with anti-social behavior in the form of weaker reciprocity, lower cooperation, lower benevolence and more bribe-oriented decisions in the corruption game. In particular, not cooperating and bribe-maximizing decisions are associated with significantly higher LSRP primary and LSRP secondary scores. Not reciprocating is associated with higher LSRP primary and being spiteful with higher LSRP secondary scores.