Abstract
Normative environments are used to regulate multi-agent interactions, by providing means for monitoring and enforcing agents' compliance with their commitments. In business encounters, agents representing business entities make contracts including norms that prescribe what agents should do. Agent autonomy, however, gives agents the ability to decide whether to fulfill or violate their commitments. In particular, when the normative space is imperfect, contracts to which norms apply may be unbalanced, making it individually rational for agents to exploit potential flaws to their own advantage. In this paper we present and analyze an approach for exerting social control within a normative environment. An adaptive mechanism is proposed that enables a normative framework to change deterrence sanctions according to the behavior of an agent population, in order to preclude agents from exploiting potential normative flaws. The system tries to avoid institutional control beyond what is strictly necessary, seeking to maximize agent contracting activity while ensuring a certain commitment compliance level, when agents have unknown risk and social attitudes. We analyze how the adaptive deterrence sanctioning model responds to different agent populations, which are characterized by predominant risk tolerance or social awareness degrees. We show that risk-averse or socially concerned populations cause lesser deterrence sanctions to be imposed by the normative system.