G8 GARRY C. MYERS 



trials before the subject learned to go directly from O to B. 

 After going directly to B three successive times it took only 

 five trials to learn to go directly from O to A again. To make 

 the next transition of the pathway (from OA to OB) it took 

 six trials. Hence, while primacy tends to persist throughout it 

 decreases somewhat with time. 



The modification in the learning is worthy of note. The 

 excited, random movements so obvious in the earlier new situ- 

 ations which successively appeared with the alternation of feed- 

 ing places, decreased as the experiment progressed. On the 

 other hand the responses to the later situations were character- 

 ized by hesitancy and by attempts to use the eyes to help deter- 

 mine, before jumping over the board K, which was the right 

 way to go. To illustrate, on the twelfth feeding which was 

 the second feeding of the third day, after having gone from 

 A to B eight successive times, he paused at K at a point on a 

 straight pathway to A from O, raised his head high enough to 

 look over K, then turned and jumped over K near the partition 

 into the B side. Likewise at the third and fourth feeding of 

 the fifth day, after making two wrong trips to B, he paused at 

 K, looked over and turned and jumped into A. His pathway, 

 moreover, instead of being in a straight line from O to B diverged 

 a little toward the partition, and on the succeeding trip diverged 

 still more, so that the next time he jumped directly into the 

 A side but near the partition. At the fifth feeding of this same 

 day, instead of looking over K, he ran to it in line with A, paused, 

 and jumped into the A side. These gradually shifting pathways 

 seem to indicate the resultant of two antagonistic impulses. 

 It must be remembered here that although the subject often 

 took the wrong pathway from O, eventually he always found 

 the food; but in finding the food after first entering the wrong 

 side the subject never went back to O for a new start, but sought 

 the food by the shortest way he could find from where he was. 

 No doubt the transitions from one pathway to another, from O, 

 could have been completed with fewer errors if the corrections 

 could have been made from O. Even then, theoretically, there 

 would have been some tendency to make the wrong trip again 

 and to return to O rather than to go to the food directly. In 

 any event, the mere recency of the appeal of the stimulus where 

 it once was, does not so much account for the tendency to con- 



