Equivalence is defined by three properties, reflexivity, symmetry and transitivity (Sidman and Tailby, 1982 ). The introductory chapters to the present thesis review the literature on equivalence. The findings from this review show that equivalence, whilst present in the matching-to-sample performance of children over five-years of age and adult subjects, is absent from the Matching-to-sample performance of animals.
The experimental chapters in this thesis present the results from developmental studies of equivalence that were designed to answer three questions. (1) can equivalence be demonstrated in children of any age? (2) Is equivalence related to the child's verbal behaviour ? (3) Can language-training procedures benefit those children who fail to show equivalence relations?
The results showed that equivalence was present in the matching-to-sample performance of four-year old children. In contrast, equivalence was shown by only half of the three year old children. Equivalence was, for the most part,
absent from the matching-to-sample performance of two-year old children.
It appears from these results that equivalence has a developmental sequence . A further analysis revealed that success on the equivalence tasks was related to the ways in which the children spontaneously labelled the stimuli. It was also found that children who fail tests for equivalence benefit from language-training procedures; in the present study, this involved teaching the children to label the
correct sample-comparison pairs during the matching-to-sample task.
The results are consistent with a growing body of literature that has found major differences between the performance of verbally-accomplished humans and animals on operant tasks (cf. Lowe, 1979; 1983) and point up the importance of
language in the control of human behaviour.