92 K. S. LASHLEY 



ing process. But we can not accept as readily the view that cere- 

 bral injury results in an improvement in learning ability. What, 

 then, is the cause of the difference between the normal and oper- 

 ated groups? The first explanation suggested is that the problem 

 box used in some way favored the injured animals. 



Observation of the animals during training indicates that the 

 double-platform box does so favor the operated animals. When 

 the normal rat encounters a low obstacle in his path he is very 

 apt to jump over it. In some preliminary tests the platforms 

 used were only 2 cm. broad. Normal animals failed to learn the 

 problem in this form because they almost invariably jumped over 

 the platforms and so failed to trip the latches. Platforms 5 cm. 

 in width were then substituted for the narrower ones and with 

 this modification the normal animals were able to learn the prob- 

 lem, although they still frequently leaped over the platforms. 

 During the experiments with the operated animals I gained the 

 impression that they leaped the platforms less frequently than 

 normal animals, but did not collect statistics on the point. A 

 crucial test of it is furnished, however, by a comparison of the 

 learning rates of paretic animals with those of the remainder 

 which showed no motor disturbance. The averages of paretic 

 and non-paretic animals are the following: 



Trials for learning 



Paretic animals required 62.6 



Non-paretic animals required 87. 1 



From this it appears that a definite paresis is advantageous for 

 learning the problem. Animals with motor disturbance are 

 more apt to trip the catches by chance than those without, and 

 they therefore have a better chance to learn. A part of the ap- 

 /parent superiority of the operated animals may thus be ascribed 

 with certainty to a lack of vigor. But how much of the total 

 difference between normal and operated animals is due to this? 

 The data at hand do not permit of a certain answer to the ques- 

 tion, but various lines of evidence point to the conclusion that 

 irrelevant factors at least did not change a real inferiority of the 

 operated animals into an apparent superiority: that the normal 

 and operated groups are most probably equal in learning ability. 



