A Framework for Emotions and Dispositions in Man-Companion Interaction - Coverbal Synchrony in Human-Machine Interaction - page 104

by-product, but as an action-managing affect factor of companion
functionality. Sloman (2010) argues that the complexity of companion
features as such should be discussed, but so should the quality and
complexity of the requirements. His starting point is the presentation
of an illustration of possible interaction flows (world knowledge), with
which a companion can compare the respectively current behavior and
therefore knows “ what they have done, what they could do, what they should
not do, why they should not do it, what the consequences of actions will be,
what further options could arise if a possible action were performed, how
all this relates to what another individual could or should do, and can also
communicate some of this to other individuals ” (p. 180). This would create
some sort of situational awareness of the companion, not necessarily
a consciousness (that would relate to a subjective experience). At
this point, it is apparent that the goals that companion technologies
have set for themselves can only be achieved with patient scientific
work, because the euphoria about the thinkable and desirable system
features of companions should not lead to the wrong conclusion that
the necessary formal descriptions of these system features have yet
been solved and if, in a rudimentary fashion, only relates to very
special, mostly simple cases. In this context, Sloman (2010) initially
views two very narrowly defined “target functions” that will become
relevant for companions in the near future, and that can be assigned to
companion technologies, i.e. “engaging function,” which mostly refer
to the quality of the interactions that entertain, draw attention, are fun
or are just interesting and “enabling functions,” which support users
with regard to their goals, motives and intentions. The latter can help
to solve various everyday problems, since they provide information,
teach, organize the user's social and physical environment and enable
the user to participate in society. Sub-functions of these target functions
are currently offered by existing companion-like assistance systems,
which are therefore very helpful as an inspiration for our visions of
future technologies.
Beverly Park Woolf from the Department of Computer Science at
the University of Massachusetts leaves no doubt: “ If computers are to
interact naturally with humans, they must express social competencies and
recognize human emotion .” Using the example of tutorial companion
systems (CS), she shows that the sensory capturing of dispositions such
as boredom, interest, and frustration by companions makes the tutoring
functions of learning supports significantly more effective, increases
motivation, and reduces adverse emotional states such as frustration,
anger, or fear (Woolf, 2010, p. 5). The companions that Woolf refers to
as social tutors capture emotional and dispositional responses during
the learning process by measuring posture, movements, grip strength,