TRANSITIVE INFERENCES USING A NONVERBAL FORM OF PRESENTATION

Transitive inference is the ability to derive a conclusion (B > D) bet ween nonadjacent items from the overlapping premises (A > B, B > C, C > D, D > E) of an underlying series. Traditionally such tasks were pre sented verbally but growing interest in animal cognition led to develo pment of a nonverbal version of this task. Therefore the premises are converted into simultaneous discrimination tasks A + B-, B + C-, C + D -, D + E- and subjects are expected to choose transitively item B in t estpair BD. The article reviews novel investigations with humans and a nimals using a nonverbal form of presentation. Transitive inference co mpetences are discussed within the scope of conditioning models and al gebraic neural net-works. Furthermore the relevance for everyday behav ior and social evolutionary behavior is discussed.