Four-pair same-different differentiation and transient memory in dogs.

In our previous study (Brown and Soltysik 1971) we investigated the same-different differentiation between auditory compounds in a threepair design with dogs (Chorqzyna 1967). In this procedure, reward was presented for responses (CRs) to the second member of a positive pair, consisting of two different tones, while no reward was given for two negative pairs, consisting of like tones. It was found that when the values of tone stimuli were held constant during training, the differentiation could be solved in part on the basis of information, concerning the occurrence of reward, present in the first member of a tone pair. This was demonstrated by an analysis of the paradigm itself, and was supported by the appearance of non-CR behaviors correlated with the first tones. It was concluded, therefore, that for the study of recent memory with the same-different task (Konorski 1959), a necessary condition is the complete ambiguity of first tones with respect to probability of reward. The present study was conducted to explore delayed responding in the same-different differentiation with a four-pair design. Two tone frequencies, low (L) and high (H), were presented in the following pairs: L-H, H-L, L-L, and H-H. In a go-no-go differentiation procedure, res-