This is a very interesting paper. The primary goal is to make a computer model of how and when emotions are made.

The authors suggest that emotions mainly come from the interruption of an important task. This interruption has survival value by focusing attention on critical tasks. They then present very brief examples of some IF-THEN rules showing what they have coded in PRISM. Their ideas are interesting, but so are old FORTRAN programs that could print out “I love you.” No evidence is even hinted at that ties their ideas to reality. Many problems arise; e.g., they say “[this] demonstrates the close interconnection of the cognitive and emotional system.” Nothing has been “demonstrated]” Talking about concepts does not make them true. Concepts may be interesting and yet be wrong. There is no discussion of the subjective nature of emotion, or how we are to deal with the “self-awareness” problem. Overall, the paper is full of interesting ideas and should be carefully studied. But it is NOT a scientific advance of knowledge nor of demonstrable facts. It is not clear why the study of emotion is an important topic in computing.