Verbal segments and multiple-pair training in relational responding under a transposition paradigm in children

Several studies have shown that the use of conventional or unconventional verbal segments (VS) can facilitate the presentation of transposition responses, even in distance tests. In those studies: A) the effect of limiting the use of VS during task resolution was not systematically evaluated; and B) only two stimuli were used during training, causing the VS to become redundant. Starting from the hypothesis that the consistency between the VS and the relational property of the stimuli could affect the functionality of the VS, the present study aimed to compare the functionality of three types of verbal segments: a) conventional-consistent (CC); b) unconventional-consistent (UC); and c) unconventional-inconsistent (UI), with 1 and 3 pairs of instances training, in a transposition task. Fifty-six children between 8 and 9 years of age participated. They were exposed to a computerized task with varying stimuli in the size dimension, as well as near and far tests. The findings of the present study suggest that: 1) CC verbal segments only became relevant, as facilitators, when more than one stimulus pair was presented in training; 2) the use of UI resulted in an interfering task that significantly limited relational responding under 3-instance-pair training; and 3) Regardless of the number of pairs of instances in training, participants trained with CCs obtained significantly higher percentages of correct answers in Test than those trained with UCs. The relevance of verbal segments and training variability to the establishment of transposition responses is discussed.