A Framework for Emotions and Dispositions in Man-Companion Interaction - Coverbal Synchrony in Human-Machine Interaction - page 124

to long-term parameters (e.g. mood, action tendency) and/or their
reciprocal effect is relevant for predicting emotional behavior. The
technical implementation of COMPLEX is shown in Figure 4.
The theoretical basis for the appraisal process currently used in
COMPLEX is the OCC model (Ortony et al., 1988). It is based on
the assumption that (discrete) emotions are the direct outcome of
an individual appraisal process, which appraises an event and/or
an action based on three aspects: (1) consequences of the event (for
one's own goals), (2) appraisal of the action on the basis of individual
standards and (3) certain aspects of objects. These three aspects are
further differentiated by the idea of several agents involved in the
interaction because the relationship between agents must also be
taken into account in the appraisal process. For example, if Person A,
who is friends with Person B, fails an exam, which is important to the
goal “graduation”, the model would generate the emotion 'pity' for
Person A (based on the positive relationship and because an important
goal was negatively impacted). For Person C, who has a negative
relationship with Person A, the OCC model would, however, predict
the emotion 'gloating' .
To implement these appraisal processes in COMPLEX, it is first of
all important to map the individual variables (events/actions, goals,
etc.) and their interplay:
Relationships between agents are described by a value ranging
from -1 to +1 (rel Ag1, Ag2 = [-1, 1]). The “1 reflects a maximum negative
and +1 reflects a maximum positive relationship. The relationship is
Figure 4. Overview of the COMPLEX model.