...clear.
One reviewer noted that this opens up the possibility of false expectations on the part of the student. This is particularly tricky with emotion modeling because of the moral issues involved. On the other hand it might well prove true that expressions such as ``Even though I am just a simple computer program, in my own small way I am happy for your success,'' may be useful without being misleading. That human-computer interaction seems to be inherently social is empirically supported by the work of Nass, and others, at Stanford. Like the indications that users seem to prefer flattery (at least in the short term), even when they know it is frivolous, it may prove true that a minimally socially responsive tutor may be an end in iteself (c.f., [Nass & Sundar1994, Nass & Moon1994, Elliott1994]).
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...moods.
More specifically, these are specified as variables which (1) alter an agent's interpretation of situations, (2) are volatile, (3) are bipolar in nature, (4) are not dispositional, so that they naturally return to (agent-specific) default values over time, and (5) may be the result of prior affective experience.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

Clark Elliott
Wed Dec 17 18:41:50 EST 1997