Artificial Emotions -- Good Bye Mr. Spock!

The question of implementing emotions in robots is twofold: on the on hand it should be verified whether such an effort is valuable, and on the other it should be determined whether the implementation is feasible. The answer to the first question seems easy: besides and beyond the reasons of pure intellectual curiosity and scientific research, emotions should be studied and implemented if the overall behavior of such robots is better than their unemotional counterparts with respect to behaving efficiently in a real world environment. Two diverse opinions have emerged in the previous discussion. One, due to McCarthy, asserts that emotions will introduce obstacles in the communication among robots and human beings [6]. On the other hand, Minsky sustains the opinion that it is impossible to implement intelligence without emotions [7]. In this paper we analyze these perspectives, discuss a possible way to approach the topic, and provide an architecture to implement emotions, which has sh..