FUNCTION OF VIBRISSAE IN BEHAVIOR OF WHITE RAT 57 



ti"ue path chanced to be the middle one. But at the same time 

 each rat had pecuHar ways of its own: some usually turned to 

 the right first, some to the left; the more active rats like "H," 

 group (a), dashed off immediately into any opening, while others 

 always hesitated. 



Habit was most marked as a sort of circular reaction and in 

 many cases, like those governed by position, was easily estab- 

 lished. "A's" nth trial for instance, was one where he went 

 back and forth as fast as possible from one alley to another, the 

 adjacent one, as if he were wound up and literally could not 

 get away. This was repeated 22 times before he finally took the 

 third alley which was the right one. 



The first choice made on a given day was likely to be a per- 

 sistent one. The path which was the right one pre\iously had 

 some attractive power. This was more noticeable in the trials 

 of a single day than from one day to another; yet in some in- 

 stances, a successful association seemed to be carried over and 

 influenced the first choices of the following day. 



There wa's little difference in action between animals with 

 vibrissae and animals without. The latter, however, "nosed" 

 the fioor more and kept closer to the sides where they could 

 use their noses. 



Errors dropped out little by little, the time decreased even 

 for animals in which you could observe no signs of discrimina- 

 tion, and the percentage of right choices was very soon above 

 the others; but after 150 trials although the errors and time 

 had steadily decreased, so that an animal often ran several days 

 without error, this behavior was not a consistent one and could 

 not be extended to five days. Most of the animals gave clear 

 evidence, so far as observation could show, that the difference 

 in the pathways was sensed in some way but they did not make 

 any consistent use of the power. In reality they could run in 

 and out of two alleys and get in at a third in less time than they 

 could do so by stopping to discriminate. From a human point 

 of view there was no reason why they should. The electric 

 apparatus was then put in and the rats were given a shock 

 if they went to the end of the wrong pathway. Most of the 

 animals could have made 60% instead of t,2,% of right choices 

 before the electricity was used but the active signs of discrim- 

 ination were lacking. The disturbance caused by the punish- 



