THE EFFECTS OF EQUIVALENCE CLASS-STRUCTURE ON TEST PERFORMANCES

Nodal distance specifies the number of nodes that separate any two stimuli in an equivalence class. Directionality of training specifies the sample and comparison functions that are served by the stimuli in a class. These parameters define the formal structure of an equivalence class. Four experiments showed that directionality of training influenced likelihood of class formation and the transfer of responding between stimuli in a class. Eight experiments showed that some test performances were inverse functions of the nodal distance, whereas others were direct functions of nodal distance. The effects of nodal distance were transient when only one response was trained to a stimulus in a class and when the effects of between-class contingencies prevailed during a particular test. They were maintained when at least two responses were trained to some stimuli in a class, and when the effects of between-class contingencies were minimized. Effects of nodal distance thus appear to be imparted permanently during the establishment of baseline relations. Their influence on test performance depends on the effects of other contingencies that can overshadow the effects of nodal distance. Test performances, then, reflect the interactive effects of nodal distance and the parameters of training and testing. The data showed that the stimuli in an equivalence class were only partially substitutable for each other. Thus the relatedness of the stimuli in an equivalence class was an inverse function of the nodal distance.