Emotions and intelligent choice: feasibility, seriousness, and admissibility

Abstract

Emotions and rationality have often been treated as opponents, with emotions taking on the role of villain and rationality taking on the role of heroine. Recently, however, the positive role of emotions has received considerable attention, and resulted in the elevation of emotion to rationality's equal. With emotion's change of status, artificial and computational intelligence researchers have begun to model and use emotions as a constructive portion of designing machines and describing humans. Building on the idea that emotions are generated by past actions (or, at least by the consequences produced by past actions) and are used to determine future actions, it is desirable to create a computational framework capable of representing these dynamics. Employing the hypothesis that some aspects of emotional state result from positive influences and other aspects result from negative influences, we can determine conditions when an action is a feasible possibility, a serious possibility, and an admissible choice.