Configurable Appraisal Dimensions for Computational Models of Emotions of Affective Agents

Abstract

In this paper we introduce the concept of configurable appraisal dimensions for computational models of emotions of affective agents. Configurable appraisal dimensions are adjusted based on internal and/or external factors of influence on the emotional evaluation of stimuli. We developed influencing models to define the extent to which influencing factors should adjust configurable appraisal dimensions. Influencing models define a relationship between a given influencing factor and a given set of configurable appraisal dimensions. Influencing models translate the influence exerted by internal and external factors on the emotional evaluation into fuzzy logic adjustments, e.g., a shift in the limits of fuzzy membership functions. We designed and implemented a computational model of emotions based on real-world data about emotions to evaluate our proposal. Our empirical evidence suggests that the proposed mechanism properly influences the emotional evaluation of stimuli of affective agents.