TRANSFER OF RESPONSE IN THE WHITE RAT 29 



We have already called attention to the fact, that with the 

 series of choices required, chance would account for 50 per 

 cent of correct responses and simple alternation between left 

 and right would account for 70 per cent of correct responses. 

 We have also called attention to the fact that the records 

 show a tendency toward simple alternation in the case of the 

 sound stimulus, but that there was no consistency at all in 

 maintaining such an alternation. We are led to conclude, 

 therefore, that whatever gain was made by those animals 

 trained with the sound stimulus was due to a partial perfecting 

 of the simple alternation position habit. In the case of 

 those which learned the positive response to the light stim- 

 ulus, since there is no direct evidence in the records of such a 

 position habit having been acquired while learning the light, 

 we would be led to believe that the light stimulus itself 

 became the increasingly dominant factor right from the 

 beginning of the learning process. However, when those 

 same animals were transferred to sound, the records immed- 

 iately show, as we have pointed out, that such a position 

 habit had been partially formed. Thus the evidence seems 

 conflicting. However, the conclusion seems warranted 

 that chance and simple alternation, imperfectly carried out, 

 will account for 60 per cent of correct responses when using 

 such a series as was used in the experiments with positive 

 response. Since such improvement is made in the first 300 

 trials any improvement beyond that amount, whether 

 during the first 300 trials or later, must be accounted for in 

 other ways. Thus Group I required only 120 trials with the 

 light stimulus to surpass 60 per cent. Group II needed only 

 180 trials to surpass 60 per cent. Our conclusion must be 

 that the light stimulus began to be effective in both cases 

 not later than during the second 100 trials and reached its 

 maximum influence at the end of 500 trials. 



But we still have the question why it required those 

 trained with the light only 120 and 180 trials to reach an 

 efficiency of 60 per cent whereas it required those trained 

 with the sound 300 trials to reach the same point. We have 

 already found that the sound stimulus had no influence 

 since it could not be localized. We have, therefore, to 

 take account of the light only. In. one case it was present 



