TRANSFER OF RESPONSE IN THE WHITE RAT 49 



(e) Attention has been called to the unusual form of the 

 majority of the individual learning curves of animals 

 learning the negative response. Contrary to the nature 

 of the ordinary animal learning curve the greatest improve- 

 ment does not take place in the earlier trials, but there is 

 at first a period of little or no progress and later a period 

 of rapid improvement with a quick perfecting of the re- 

 sponse. In form these curves come more nearly into agree- 

 ment with Swift's curves on Learning to Toss and Catch 

 Balls. *As far as the writer is able to discover, the curves 

 given in this paper are the first animal curves of this nature 

 to be reported. Going on the assumption that perhaps the 

 difference was due to the fact that all the evidences of 

 learning were not being recorded, an attempt was made to 

 correct this failure by taking note of " hesitations " and 

 " retardations " in the animals' actions which could evi- 

 dently be assigned to the stimulus as cause. But even this 

 did not " correct " the form of the curves. As far as could 

 be determined " errors " or " choices," with modifications 

 of the latter in the form of " hesitations " and " retarda- 

 tions of movement," were the only criteria which could be 

 made use of in measuring the progress of learning or the 

 relative rate of learning. Unless we posit some change in 

 the organism which is produced by the various trials but 

 which does not manifest itself extrinsically in measurable 

 behavior until a certain degree of strength has been reached, 

 and then manifests itself by taking almost complete control 

 of the animal's activity' in that situation, we are at a loss 

 to account for the difference in the rate of progress of 

 learning. That would mean that in certain kinds of prob- 

 lems we might expect learning curves of this kind or, in 

 other words, that the shape of the learning curve depends 

 in part upon the nature or kind of problem. 



In contrast with these curves for the learning of the 



* Note: Professor Yerkes, of Harvard, reports in his monograph on the Mental 

 Life of Monkeys and Apes (Behavior Monographs, Volume 3, Number 1, 1916, page 

 68) that one of his animals (Julius) showed the type of learning curv^e found in the 

 work reported in this paper and claims that never before has a curve of learning like 

 this been obtained from an infra-human animal. However, his work was done 

 subsequent to the work reported in this paper. The work reported here was done 

 during 1913 and 1914, but, owing to unavoidable delay has not been published 

 until now. 



