The literature on the Orbito Frontal Cortex (OFC) is growing rapidly, implicating OFC in numerous cognitive functions (response inhibition, valuation, credit assignment etc.,) and neuronal disorders. Better understanding of these functions is crucial in many domains studying human cognition including Artificial Intelligence. However, a number of ideas that explain OFC function and the data sets used to explain these ideas have been questioned (Stalnaker et al., 2015). This diversity in the proposed OFC functions and the attempts to invalidate them, simultaneously highlight the necessity and importance of designing experiments that allow to challenge the validity of these ideas. Moreover these ideas might have to be studied alongside each other, rather than individually, demonstrating their role in the emerging behavior. We propose a framework to study a variety of such functions in the form of the behavior of an artificial agent in a virtual experimentation environment, the video game Minecraft. This environment has been exploited before to create scenarios in the domain of Artificial Intelligence (Strannegard et al., 2018) and we develop an adaption of it to help test functional hypotheses in the field of neuroscience. Precisely, we attempt to model certain functions of the lateral and the medial OFC (lOFC and mOFC) that have been widely studied and reported (Noonan et al., 2010, 2011, 2012). We refer to the implications of lOFC in pavlovian responses in the context of multiple stimuli and those of mOFC in making a decision from a reduced number of choices filtered by lOFC. To observe how these roles of OFC fit in the context of emergent behavior, the framework also utilizes functional description grounded in the understanding of the roles of other regions of PFC like Anterior Cingulate Cortex (ACC) and those which form the primitive sensori-motor loops with the basal ganglia. Numerous testable predictions can be made from such a framework that help study each of these regions in detail. The framework is expected to be generic with same learning and decision-making mechanisms operating at different functional levels. Studying such frameworks of cognition also enables a constructive approach to a class of problems related t