Extrinsic Emotion Regulation by Intelligent Agents: A Computational Model Based on Arousal-Valence Dimensions.

Emotion regulation is an important aspect of emotional well-being that involves effectively managing and modifying one’s or other emotions. However, there is a lack of computational models to guide the design and implementation of emotion regulation by intelligent agents. To address this gap, we propose a computational model for an intelligent agent which, using an emotion representation based on the arousal-valence dimensions, facilitates emotion regulation in persons. Based on the theoretical emotion regulation model proposed by J. Gross, our model of intelligent affective agent uses a dynamic planner to select a regulation strategy to facilitate the individual to maintain their emotional balance. The agent selects the most appropriate emotion regulation strategy taking into account the specific personality traits of the individual. The effects of each regulatory strategy on a particular individual are updated by the agent according to the emotional changes detected in the individual physiological parameters in previous uses of the strategy. The proposed model of agent is of great importance to facilitate the learning and use of emotion regulation techniques for promoting emotional well-being of people.