676 WILLIAM J. ARNOLD 



tion problem. Training continued until each animal reached a criterion of 

 8 consecutive trials without an error or until a maximum of 50 trials was 

 reached. 



Results 



Only one irradiated rat reached criterion, by comparison with 8 rats in 

 the control group. Significantly more errors were made by the irradiated 

 group than by the control group. These findings indicate that the long term 

 effects of irradiation are deleterious. 



EflFects of Brain Irradiation as a Function 

 of the Interval after Irradiation 



The finding that radiation has long-term deleterious effects on discrimina- 

 tion learning led to a more extensive investigation. An experiment was de- 

 signed to determine the effects of 5,000 r brain irradiation on maze learning, 

 discrimination learning, and concept formation at intervals of 1, 3, 6, 9, and 

 12 months after irradiation. Two hundred male and 200 female Sprague- 

 Dawley albino rats, 90 days old, served as subjects. Maze learning was 

 evaluated by means of a 14-unit T-maze, and a brightness discrimination was 

 established in a five choice discrimination box. In order to test learning set 

 formation, the previously described series of oddity problems were used. An 

 increasingly rapid solution of the successive problems was taken as evidence 

 of the formation of a learning set. 



Within each sex, rats were randomly assigned to experimental and control 

 groups for each of the test intervals (1, 3, 6, 9, and 12 months) . All experi- 

 mental groups were exposed to 5,000 r brain irradiation by our standardized 

 technique, and controls were sham irradiated. At the various postirradiation 

 time intervals, the groups were tested for maze learning. Then they were 

 tested for discrimination learning, and then, concept formation. To keep the 

 postirradiation time interval the same in each group, each of the tests was 

 begun only after all the animals of a group had finished the preceding test. 



Results 



Because of a high mortality, due in part to an outbreak of respiratory 

 infection and in part to the effects of radiation, maze learning data were 

 obtained only for the 1, 3, and 6 month groups, discrimination learning data 

 only for the 1 and 3 month groups, and learning set data only for the 1 

 month group. 



The irradiated animals were consistently superior in maze learning, and 



