150 Studies in Animal Behavior 



is no apparent congruity between scratching and get- 

 ting out and being fed. But when these acts have 

 been performed a number of times in close sequence 

 we cannot assume that they will not form a con- 

 gruous association. The investigations of Pawlow 

 and his co-workers have gone far to show that al- 

 most any two acts performed at nearly the same 

 time may become associated. The salivary reflex 

 of a dog may be set up through having become as- 

 sociated with sights, sounds, stimulation of the legs, 

 or even the application of a painful electric stimu- 

 lus. How the second act influences the first after 

 the association has been formed is the important 

 consideration, and we need not be surprised to find 

 that certain responses are reinforced by acts of 

 very diverse kinds. 



"Again," says Thorndike, "if a cat is put into a 

 box, X, with two alleys opening to the North from 

 it, A and B, and if whenever it advances two feet 

 into alley A it is hit from behind with a club and 

 so runs on out of the North end of A, whereas, 

 if it advances two feet into alley B, it is given a piece 

 of meat and hit gently from in front, the cat will, 

 when put into X, be less likely to advance into A 

 and more likely to advance into B. Yet the response 

 of advancing into A produced the congruous sec- 

 ondary response of advancing further in the same 

 direction, whereas the response of advancing into 

 B produced the incongruous retreat into X." 



I think the experiment, however, is capable of a 



