Findings from studies on stimulus equivalence with human subjects suggest that different verbal processes might be related to success in equivalence tests. The purpose of the research for this thesis was to directly compare the effects of common names, different names, and different names plus relational terms on the performance of 4-5 year old children in (i) forming equivalence classes and (ii) extending these classes.
The results of Experiment 1 show that common naming can only facilitate equivalence class formation if the appropriate contextual support is offered. A history of identity training provided the contextual support for common naming to facilitate equivalence class formation for 3 out of 4 children. These results pointed out the need to investigate the effects of identity training. Subsequently, Experiment 4 was conducted, which showed that a history of identity training alone does not facilitate the emergence of the AB MTS (matching to sample)
relation. The data from Experiments 2 (different names) and 3 (different names plus relational terms) demonstrated that a history of identity training does not facilitate the emergence of the AB MTS relation when stimuli have different names. These results suggest that the effectiveness of a given procedure appears to be dependent upon a subject's naming of the stimuli.
It seems that the AB and AC verbal chains functioned as intraverbals for one of the subjects in Experiment 2, resulting in the A,B,C names being grouped together which facilitated the formation of equivalence classes with different names. In addition, the results from Experiment 5 suggest that the establishment of the AB and AC relations before testing for equivalence and the presentation of baseline trials during test sessions may be the contextual support required for different names to facilitate stimulus equivalence.
The results from 2 of the 3 subjects in Experiment 3 show that the use of different names plus relational terms can be more effective than the use of verbal chains with different names in the establishment of the AB relation. However, no evidence has been provided as to the facilitative effect of these relational terms in the formation of equivalence classes.
The results of Experiments 1 and 2 show that common naming seems to be a particularly efficient means of facilitating the expansion of equivalence classes. Finally, the data from the present thesis adds support to those studies which suggest that language is related to stimulus equivalence. They also suggest that there is not just one type of equivalence corresponding to the mathematical concept but that a multiplicity of behavioural processes exist, all of which may lead to success on equivalence tests but may differ in their efficacy in
extending the stimulus classes.