DISTRIBUTION OF EFFORT IN LEARNING IN WHITE RAT 7 



practice the rats were not left idle in their cages, but were per- 

 mitted to exercise once a day in the maze when feeding. It was 

 the only method at hand to avoid the evils that arise from close 

 confinement, of which there are a great many. The most serious 

 one is the rapid increase in weight that usually occurs after a 

 period of experimentation. A rat, having been once restricted 

 as regards food, has a tendency to overfeed and to take on fat. 

 Several days before the retention tests were made the amount 

 of food w^as still more restricted, and conditions similar to those 

 maintained when learning was in progress were established. 



RESULTS WITH THE LATCH-BOX 



As a rule, as soon as the rats entered the hood, they made 

 directly for the door of the latch box, the more timid approach- 

 ing cautiously. Having been fed previously for two days in 

 the problem box, and finding their way obstructed by the closed 

 door, they usually ran around and over the problem box and 

 climbed inside the hood, but eventually attacked the door. Their 

 attack was often violent, and all methods imaginable were tried 

 before success w^as attained. As a result, the first solution of 

 the problem was made in complex and in unique ways. But 

 these w^ays were soon to disappear, and with them the sharp- 

 ness and the suddenness of action which appeared in the move- 

 ments of the early trials. With their disappearance a noticeable 

 modification occurred, movements became more precise, indi- 

 cating that integration was in progress. 



To print the results of the daily records of each rat is im- 

 possible and accordingly only the average results are here pre- 

 sented. These are given by distribution curves and by tables. 

 The former are arranged to give the number of trials and the 

 number of days each rat required to learn the problem and 

 their distribution; the latter, the averages for each trial for all 

 rats in the one, three, and five trial groups. From the tables 

 ordinary curves are plotted to show the process of the integra- 

 tion of the movements. Interesting as the records of each rat 

 would be, the important data, the averages, will show sufficiently 

 well the more important results of the experiment. 



The number of trials, or the total time consumed in learning, 

 and the distribution in the number of days before each rat 

 mastered the problem are indicated on the distribution curve. 



