The Emergent Emotion Model
|
In the Emergent Emotion Model, motivation is represented by the needs of the guide to maintain its performance and establish an acceptable condition for interaction with the user. Emotions are emergently determined by the modulators while cognition is represented by the information processes in PERCEPT, GENERATE INTENTION, SELECT INTENTION, RUN INTENTION, RECOVER as well as in the memory of intentions represented by the purple boxes in Figure 1. Basically, the guide has two built-in motivators to maintain, competence and certainty. The level of competence refers to the guide's ability to cope with the user's differing perspective about an issue whereas the level of certainty is the degree of predictability of the user interests. The user's confirmation or not of the guide's prediction about their interest, contributes to its level of certainty while the degree to which the user agrees with its argument, contributes to its level of competence. These built-in motivator values will affect the modulator values, such as arousal level, resolution level and selection threshold.Arousal level is the guide's readiness to act while resolution level determines the carefulness of the guide's behaviour. The selection threshold on the other hand, is the limit competing motives have to cross in order to become active. Arousal level is inversely related to resolution level while directly proportional to selection threshold. The interaction between the built-in motivators and the modulators produces wide variation of emotions and behaviours in the guide.
|
![]() Figure 1. The Emergent Emotion Model |
|
|
Functionally, the guide continuously reads user inputs (feedback on the degree of interest in the story and degree of agreement to the guide's argument), together with system feedback (success or failure) and the GPS information. The user gives feedback through two trackbars (one for the degree of interest and the other for the degree of agreement) that allow rating from 1 to 10. The rating of 1 means not interested in the story or total disagreement to the guide's argument, while the rating of 10 represents high interest in the story or total agreement to the guide's argument. Then, it generates an intention and stores it in a memory of intentions together with the built-in motivators. More than one intention can be active at the same time and depending on the importance of the need and its selection threshold value, one of the active intentions is selected for execution. The execution of an intention will produce a success or failure feedback into the system and recovery will be performed as necessary. The guide has three possible intentions - update its beliefs about the user's interests; adjust the story presentation; or perform storytelling. In order to execute an intention, the guide decides whether to explore for more information, to design a plan using the available information or to run an existing plan. The decision on how to perform the intention is made based on the value of its built-in motivators and modulators, or in in other words, the guide's current emotional state. A standard prompt is generated when there is no story to tell or the story for the current location has finished at which point the guide informs the user of the unavailability of any story. Planning is performed for storytelling and the extensiveness of planning depends on the guide's resolution level. In the case of update beliefs or story adjustment, usually occur when needs are high due to user's disinterest in the stories or disagreement to its perspectives, the guide will explore the database for information. The information gathered will allow appropriate changes to its beliefs about the user's interest and story topic. By doing this, it adapts its behaviour according to its internal states and the environmental circumstances. For example, if the guide's prediction about the user's interests is correct (high certainty) and the user perspective is consistent with that of the guide (high competence), the guide may experience low to medium arousal level and selection threshold with a medium resolution level. In this case, the guide can master the situation. The guide will perform planning and provide an elaborated story on the current subject. When the resolution level is high enough, it will also include its own viewpoints in the narration. On the other hand, if the guide's prediction about the user's interests is wrong (low certainty) and the user's perspective is in conflict with the guide's viewpoint (low competence), the arousal level of the guide will be very high. As a result, selection threshold will be high and resolution level will be low. It is reasonable to react quickly and concentrate on the current task. In this situation, the guide tends to give a brief story without details. |
||