Empirical determination of membership functions for stimuli comparison

A generalization of the classical problem in psychology of the modeling of stimuli discrimination is considered. When in this context fuzzy attributes are assumed, from a practical point of view the main problem to be solved is that of the construction of the membership functions with which the experts involved define that attribute. Hence, in this paper, conventional elicitation membership functions methods are briefly surveyed. A method tailored to the problem considered is designed and shown. According to this method, a real practical problem is solved. Finally, coherence measures are introduced, and some of their properties are presented and analyzed