...emotion

Preparation of this article was supported in part by Andersen Consulting through Northwestern University's Institute for the Learning Sciences. Citation appears in bib.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...faces

Although reportedly the work of Eckman might begin to address this issue.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...state.

And this, even though the goal of the developers of the package was to eliminate inflection and simply recognize the words.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...so.

Interestingly, when not imagining a situation consistent with the emotion we were attempting to express by voice, we were consistently less accurate. In addition, it appeared that the computer was biased somewhat in its errors toward the category the user was imagining, for certain pairs of categories, which might suggest an interesting study.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Emotion

Birmingham, England, August, 1993

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.