Modeling naive causality in everyday reasoning with fuzzy logic

The aim of this paper is to present a new approach to the representation and elaboration of fuzzy causal reasoning. The proposed approach is based on some results obtained by several studies on causal explanation in the field of cognitive sciences. Drawing form such results, we present a fuzzy linguistic inference called generalized equivalence that permits to represent causal relationships contained in causal linguistic explanations though fuzzy relationships between antecedents and consequents. The generalized equivalence expresses the uncertainty of the causal link in an approximate way. The proposed model can be used to represent verbal explanation containing fuzzy evaluations of variables and of the relationships among them, such as in the statementusually bad weather causes a remarkable increase in car accidents, where usually, bad weather and remarkable increase are fuzzy constructs. The generalized equivalence can be applied to fuzzy causal maps to represent the intensity of causal relationships between fuzzy concepts.