Computer simulation of social phenomena is a promising field of research at the intersection between social, mathematical and computer sciences. Agent-based technologies have been used to simulate human society and aim to provide explanation for emerging behaviours. As software agents are encoded as executable program code, their struc-ture and underlying architecture frame the way how the decision-making process is described. Current architectures mainly origin from computer science and do not make efficient attempts to describe human-like reason-ing or consider the language of non-computing domains. This can create obstacles in communication between involved researchers; hence we pro-pose here a new agent architecture based on the Theory of Interpersonal Behaviour. This theory originates from psychology, and therefore we con-sider it to be more suitable for the usage in interdisciplinary research, as it enables domain experts to encode better human-like decision making processes.