Effect of Identity versus Oddity Training on Novel Matching-to-Sample Responding after Naming
Identity and oddity matching-to-sample tasks were arranged for different groups of five-year-old children. The children were then taught to name stimuli A1, B1, A2, and B2, which had not occurred in the identity or oddity task. Two names were taught; one was used for two stimuli and the other name for the other two stimuli. Finally, matching-to-sample tests with A1, B1, A2, and B2 were carried out. These offered a choice between a comparison with the same name as the sample and a comparison with a name other than that of the sample. The children who had received identity training tended to choose the comparison with the same name as the sample. The children who had received oddity training tended to choose the comparison with the other name. The results obtained with the oddity group suggest that giving the same name to two stimuli is not sufficient for equivalence of these stimuli. The results obtained with both groups can be explained on the basis of pre-experimental abstraction processes.