Cognitive modulation of appraisal variables in the emotion process of autonomous agents

In this paper, we present a scheme for modulating the results of evaluating emotional stimuli using appraisal dimensions. The level of modulation depends on cognitive information projected from components of a cognitive agent architecture. In particular, the proposed scheme models the influence of cognition on appraisal dimensions by modifying the limits of fuzzy membership functions associated to each appraisal dimensions in real time. The computational scheme proposed is designed in the context of an Integrative Framework designed to facilitate, through input and output interfaces, the development of computational models of emotion capable of interacting with cognitive components implemented in a given cognitive agent architecture. We also describe a possible scenario to show the functionality of the proposed scheme. This scheme is aimed at modeling in autonomous agents the interaction that occurs between cognitive and affective processes in humans.