4i6 'Journal of Comparative Neurology and Psychology. 



finally, a repetition of tlaose movements. The association between ttie door 

 and the plug. lio\A'ever, did not seem perfect until after No. 6 had succeeded 

 several times in pulling the plug and in getting food. The tendency of No. (3 

 was to center his attention on the door and, after failure there, to resort to 

 the plug. This may have been due to the fact that he once got food by 

 working directly at the door. 



General Summary uf the Results of the Plug Experiment. 



In nicety of imitative behavior, the Plug Experiment furnishes 

 less satisfactory results than do some of the other experiments. This 

 is no doubt due in part to the fact that the food door and the 

 means of its opening were in different parts of the cage. The two 

 things could not well be within the range of vision at the same time. 

 In transferring attention from the door to the plug the animal 

 usually lost sight of the door. lie did not see the imitatee pull the 

 plug and at the same time see the result of the pull. In case he 

 saw the plug pulled, his eyes must follow the imitatee back to the 

 door in order to see the result. Despite this difficulty, the experi- 

 ment yielded two cases of behavior in which the influence of the 

 imitatee was sufficient to guide the behavior of the observing animals 

 to a successful issue. In the successful behavior there seemed to 

 be an element of accident. It is impossible, however, to explain 

 the conduct of either No. 6 or Iso. 2 as a case of random movement 

 and accidental success, for prolonged opportunity to solve the problem 

 in this way resulted in failure for each of them. ISTor does it seem 

 possible to think that I^o. 2 or !No. 6 repeated the movements of No. 

 5 merely from seeing her perform the act and without connecting with 

 her act the result which followed it. Each of the animals failed to 

 pull the plug after seeing it pulled, until there had been abundant 

 opportunity to see the performing animal get food. 



TABLE 27. 



Results ok the Plug Experiment. 



I. 



Number of animals used in imitation tests 2 



Cases of successful imitation 2 



Cases of partially successful imitation 



Cases of failure to imitate 



