RONALD E. MYERS 5OI 



shows that discriniination IV-ab was involved in all such instances. It was 

 at first thought that the reduction in contralateral recall might be a reflec- 

 tion of some peculiarity in this discrimination such as a too close similarity 

 between the two patterns of the pair resulting in an inordinate difficulty 

 in their differentiation. This interpretation seemed unlikely, however, 

 since no such difficulty was reflected in the initial learning curves of the 

 cats trained on this response. Furthermore one cat, Hrl, was seen to 

 transfer this response at a relatively high level. A second cat, Kiis, who at 

 first exhibited a strong positional perseveration on transfer testing, later 

 revealed a good transfer performance on testing with polarity reversal. 

 This latter case must, however, remain inconclusive. There remained for 

 clarification the depressed transfer performances of cats Bsh and Bni on 

 discrimination IV-ab. 



Reference once again to Table I shows that both Bsh and Bni had earlier 

 been taught discrimination Ill-ab through the same eye that had been 

 used for the transfer testing of discrimination IV-ab. It seemed possible 

 from inspection that the depression of transfer performance with discrimina- 

 tion IV-ab may have resulted from resemblances between its }iei^ative 

 pattern, IV-b, and the positive pattern, Ill-a, of the antecedent discrimina- 

 tion Ill-ab. Thus was it first suggested that a degree of conflict in the 

 experience of the two eyes of the chiasma-sectioned cat may result in 

 interference with interocular transfer. 



The final investigation to be reported was designed to test this hypo- 

 thesis, and, if it be true, to determine the extent to which conflict may 

 hincier transfer. Failure to resolve conflict between such monocularly- 

 learned responses would indicate a defnnte limitation in corpus callosum 

 function. 



Patterns iii-a and iii-b were constructed such that iii-a closely resembled 

 Ill-a and iii-b closely resembled Ill-b (see Fig. lo). Quite closely con- 

 flicting discriminations could then be created by making pattern Ill-a 

 positive for one response and iii-b positive for the other. Three chiasma- 

 sectioned cats were used in this study and were taught discriminaton 

 Ill-ab through one eye, and, after 120 trials of overtraining, the closely 

 conflicting discrimination iii-ba through the other eye. Such conditioning 

 was alternated back and forth between the eyes until each eye immediately 

 performed its own response correctly on each new presentation. These 

 monocularly established responses were then further stabilizeci by the 

 usual periods of overtraining (400 trials seen with each). 



The patterns of the discrimination taught through one eye were then 

 presented for the first time to the other eye for tests of interocular transfer; 



