i8 4 MIND IN EVOLUTION CHAP. 



there might be reflective imitation imitation based on 

 the perception of another's act and its result to that 

 other remains uncertain. 



What has first to be settled is the possibility of a still 

 simpler mental act learning by the perception of an event 

 and its consequence when that consequence directly affects 

 the learner. If an animal sees something done which has 

 the immediate effect of giving it what it wants, then the 

 something done falls within the sphere of interest as above 



the others who formed the association by accident. But I notice in 

 Mr. Thorndike's table that this particular cat only succe6ded in one other 

 box ; we are not told in how many it failed. In any case, Mr. Thorndike 

 seems to be in error in arguing that there was no imitation because the 

 one cat used its teeth to pull the string, and the other its claw. In my 

 experiments I found that animals frequently varied their own method in 

 this fashion. Mr. Thorndike's argument would only hold against a purely 

 mechanical and unintelligent theory of imitation. 



Another cat failed to climb up netting, though eighty times it saw other 

 cats climb up and get fed ; but as Mr. Thorndike tells us that even when 

 a piece of fish was held out through the netting he would not climb after 

 it, this result is not very surprising. If the cat would not climb to get 

 fish which he saw before him, it is not likely that he would climb because 

 he saw others do so. 



In the remaining experiment mentioned by Mr. Thorndike, two cats 

 learned to do as a third, but they did not " form the association more 

 quickly than they would have done alone." But how quickly would they 

 have formed it alone? One could only judge from the speed of other 

 cats, and nothing is more unsafe than to infer from the performance of 

 one animal to that of another. 



Mr. W. S. Small's experiments with rats (American Journal of 

 Psychology, Jan. 1900) are more satisfactory on this point. They go to 

 show (a) that when two or more rats are confronted with a difficulty 

 (e.g., to find their way into a box to get food) one will overcome it first, 

 and there will be no tendency on the part of the others to learn from him 

 as long as he is there. They will follow him, and very likely rob him, 

 but they will leave him to do the work. But (b) if the pioneer rat is 

 removed, the effect is not uniform. In several instances the remaining 

 rats seemed to have learned nothing ; but in the case of four young rats 

 called A, B, C, and D, the results seemed to favour imitation. Here rat A 

 learnt the trick first in an hour and a quarter. As long as he remained 

 in the box, no other rat seems to have made any attempt to help or learn 

 from him, but on his removal, rat B made the attempt, and succeeded in 

 two and a half minutes. Again, when B was removed, c succeeded in 

 the same time. D was not tried alone. This does, I think, suggest that 

 both B and C had learnt something from A, but Mr. Small seems to 

 think that what they learnt was merely to attack the door and not waste 

 time on the rest of the box. What is most instructive in aiding us to 

 judge the value of negative results is the total change in the behav- 

 iour of these rats as soon as the pioneer is removed. If a rat finds 

 that stealing is easier than work, he will leave his friend to work, and for 

 his own part will adhere to his stealing. There is evidently a good deal 

 of human nature in rats. , 



