I. Introduction
Appraisal theory is one of the more influential perspectives on emotion and arguably the most fruitful source for those interested in the design of intelligent systems, as it emphasizes and explains the connection between emotion and the symbolic reasoning processes. Indeed, the large majority of computational models of emotion stem from this tradition (for recent reviews, see [1], [2]). In appraisal theory, emotion arises from patterns of individual judgment concerning the relationship between events and an individual's beliefs, desires and intentions [3]. These judgments characterize the personal-significance events in terms of a fixed set of abstract criteria, called appraisal variables, and include considerations such as whether events are congruent with the individual's goals, are they expected and are they controllable. Patterns of appraisal trigger specific emotions. For example, a surprising and uncontrollable event might provoke fear. Although most appraisal theories (and models based on them) emphasize the immediate emotional reactions to events, emotions unfold and shift over longer periods of time and some theories (and models) posit that these changes are mediated by coping strategies-e.g., planning, procrastination or resignation-which modulate emotional responses through a continual cycle of appraisal and re-appraisal [3 p. 127, 4].