266 Animal Intelligence 



The reduction of all learning to making and rewarding 

 or avoiding and punishing connections between situation 

 and response allows changes in intellect and character to 

 be explained by changes in the neurones that are known 

 either to be or to be possible. I have elsewhere sketched 

 one such possible neural mechanism for the law of effect. 1 



On the contrary, imitation, suggestion and reasoning, 

 as commonly described, put an intolerable burden upon 

 the neurones. To any one who has tried to imagine a 

 possible action in the neurones to parallel the traditional 

 power of the mere perception of an act in another or of 

 the mere representation of an act as done by oneself to 

 produce that act, this is a great merit. For the only 

 adequate psychological parallel of traditional imitation 

 and suggestion would be the original existence or the gratui- 

 tous formation of a connection between (i) each neurone- 

 action corresponding to a percept of an act done by another 

 or to the idea of an act done by oneself and (2) the neurone- 

 action arousing that act. It is incredible that the neurone- 

 action corresponding to the perception of a response in 

 another, or to the idea of a response in oneself, or to the first 

 term in an association by similarity, should have, in and 

 of itself, a special power to determine that the next neurone- 

 action should be that paralleling the response in question. 

 And there is no possible physiological parallel of a power 

 to jump from premise to conclusion for no other reason 

 than the ideal fitness of the sequence. 



X 



SIMPLIFICATIONS OF THE LAWS OF EXERCISE AND EFFECT 



There has been one notable attempt to explain the facts 

 of learning by an even simpler theory than that represented 



1 In Essays Philosophical and Psychological in Honor of William James, 

 PP- 591-599. 



