SOiME EXPERIMENTAL RESULTS 241 



SUMMARY. 



To put these points briefly together, the experiments 

 suggest : 



1. That animals learn by attention to a simple sequence 

 of events. It is easiest for them to learn if the first event 

 is an act of their own, and the second a result of that act, 

 which gratifies or hurts them. But even in those cases, 

 their method of learning does not ordinarily, in the 

 instances which I have observed, conform to the notion of 

 the gradual growth or inhibition of a habit. It conforms 

 rather to the rise of an idea, at first perhaps dimly grasped ; 

 then clearly seen ; for a while waveringly held, but soon 

 definitely established. 



2. For reasons that have been given, the basis of what 

 was learnt was, in several cases, most probably the per- 

 ception of what was done by the experimenter, and its 

 result. This, however, requires further proof. 



3. Whether by perception of what was done by another, 

 or by noting the results of their own actions, it seems fair 

 to say though on this point, also, further evidence is 

 required that what the animal learnt to do was, in some 

 instances, though not in all, to effect a certain perceptual 

 change as a step to securing food. Their behaviour can- 

 not be described in all cases as a uniform motor reaction 

 to a perception but rather suggested a combination of efforts 

 to effect a definite change in the perceived object. Such a 

 direction of action to an external change, I call a practical 

 idea. And the correlation of such an idea with a remoter 

 end, I call a practical judgment. 



4. This "idea," however, does not represent any analysis 

 of what is perceived. Animal perception as compared with 

 human perception would seem, by the use of it, to be 

 crude; and the ideas derived from it are no less crude. 



5. There is no natural tendency to learn by perception ; 

 still less to <c reflective " as distinct from " simple " 

 imitation. 1 



1 Mr. Small says, " It is not impossible that a form of imitation, 

 involving the higher associative processes, might be demonstrated if it 

 were possible to direct the attention of the rat to the actions of the other 

 rat, while retaining undiminished the affective basis for action" (p. 163, 

 op. cit.\ Except that the animals used to watch, not the actions of another, 



