EFFECT OF DELAYED FEEDING UPON LEARNING 53 



The twelve animals were divided into two groups. Six of 

 these had to learn the problem by the usual method of immediate 

 feeding. Six others were required to learn by the method of de- 

 layed feeding. The animals chosen were laboratory pets and 

 exceedingly gentle. All twelve were bred from stock which had 

 been in the laboratory for a long time. 2 The twelve animals 

 were distributed at random into the two groups before the first 

 trial. The only precaution taken was to see that each group con- 

 tained three young animals and three of the older ones. The 

 early trials (table 1) show that the initial ability (after the first 

 trial, which need not enter our records because of the fact that 

 it was not until after the first solution of the problem that any 

 difference in procedure was introduced) of the two groups was not 

 very different. All twelve animals were allowed to get their 

 food in the box before the first trials were given. During this 

 preliminary habituation stage the opening at b was closed so 

 that the explorations of the animals were confined to the inside 

 of the problem box proper. A hinged door in the top of the box 

 permitted the animals to be lifted in and out. In the regular 

 trials the problem box was banked up with sawdust on all four 

 sides to a height of four inches. At all times (and with both 

 groups) during the tests the lid to the food box was left on. 

 After a given animal, working by the immediate feeding method, 

 had scratched away the sawdust and entered 6, the door was 

 closed and the food box immediately opened. It was always 

 opened by the time the animal could pass up through c. The 

 rats were allowed to eat for five seconds and then they were 

 lifted out and taken back to their living cages. Only one trial 

 per day was given. Exactly the same method was adopted for 

 the group working by the delayed feeding method except that 

 in this case the lid to the food box was held down for thirty seconds. 

 The behavior of the animals working under the ordinary con- 

 ditions offers nothing worthy of comment. The behavior of the 

 group whose feeding was delayed for thirty seconds presents an 

 unusually difficult problem to those who hold that the getting 



2 1 wish to thank Dr. Helen Hubbert for supplying me with these animals. 



