242 MIND IN EVOLUTION CHAP. 



It may be added that in spite of much difference in the 

 way of going about a thing, there appeared to be no 

 essential difference in capacity to learn between the dogs, 

 elephants, cats, and otter. The difference between in- 

 dividual dogs, for example, appeared much more marked 

 that those beween the dog Jack and the elephant Lily. 1 



but my motions, this sentence expresses in a few words the point on 

 which success or failure in the above experiments appeared to turn. 



1 The problem of perceptual learning has been the subject of a large 

 number of laboratory experiments since the above was written, many 

 of which gave negative results. It must be remarked, however, that 

 the question is not whether the lower animals in general can learn through 

 perception, nor whether the higher animals normally use this form of 

 intelligence, but whether some of them are on occasion capable of 

 so doing. It is therefore a case in which the affirmative instances are 

 decisive as against the negative, and in a number of cases the affirm- 

 ative appears to be clearly made out. To begin with monkeys, 

 Kinnaman ("Mental Life of Two Macacus Rhesus Monkeys in Captivity," 

 American Journal of Psychology, Vol. XIII.), while associating his 

 general account of their methods with that of Thorndike's cats, shows 

 clearly that the female learnt from her male companion to pull out a 

 plug in which she had failed when by herself, and Dr. Haggerty 

 (" Imitation in Monkeys," J.C.N., Vol. XIX.) showed that of eleven 

 monkeys which she used all but two exhibited imitative behaviour (p. 

 434). Her remarks upon the nature and stimulus of imitation are very 

 instructive. In the first place she points out that in the most successful cases 

 the imitator responds " by attacking a particular object." He does not 

 necessarily repeat the " movements of the imitatee in detail. The impulse 

 seemed to be to do something to the object, and the imitating animal 

 used his hands and teeth interchangeably " (p. 439). This accords with 

 my experience. It is against a mechanical interpretation of the imitative 

 act and favours its purposive character. Further, " there is evidence to 

 show that in certain cases the behaviour of the animal, unaccompanied by 

 any profitable result, is not sufficient to produce imitation." Our author 

 concludes : " Thus the facts would indicate that not only the act of the 

 animal but also the profitable result of that act was a necessary feature 

 in producing imitation" (p. 441). 



Rapidity of learning comes out markedly in Kinnaman's experiments 

 in relation to complex acts such as opening a door by turning a button, 

 lifting a lever, and so on. In several of these cases the time falls plumply 

 after the first trial in a relation of ten to one or more. 



Coming lower than monkeys, we have Mr. C. S. Berry's studies of rats 

 and cats. In the case of rats Mr. Berry concludes, like Dr. Haggerty, 

 that the result is of importance. The untrained rat learnt the way out 

 from the trained one, but " it was not until the trained rat had got out a 

 few times that the untrained rat began to follow him." At the beginning 

 they showed "no particular tendency to take note of each other's moves." 

 ("The Imitative Tendency of White Rats," J.C.N.P., Vol. XVI., especially 

 p. 359). His experiments on cats ("An Experimental Study of Imitation 

 in Cats," J.C.N.P., Vol. XVIII.) gave very similar results. Imitation 

 occurred and once again Mr. Berry interpreted it as due to the perception 

 of the result rather than to the perception of the performance of the act 

 itself by the trained animal. " In many cases I think it is not so much 



