IV. TRANSFER OF TRAINING BETWEEN A MAZE PARTIALLY 

 LEARNED AND ONE COMPLETELY LEARNED 



We have seen that there is always transfer of training 

 when both mazes of a pair are completely learned, and 

 that the transfer is positive. The inquiry may be made 

 as to whether there is transfer of training when the first 

 maze of the pair is only partially learned and the second 

 completely learned. 



In order to answer this inquiry, groups of rats were 

 transferred from one maze to another after two, four, 

 eight, or sixteen trials in the first of the two mazes, and 

 also after the complete mastery of the first maze. The 

 groups consisted of from seven to twelve rats, with an 

 average of nine. The mazes used were the ones in which 

 the greatest degree and the least of positive transfer of 

 training had occurred in the last experiment considering 

 the average of all the groups — mazes E and D respectively. 

 The results are presented in tables 33 to 3S, and graphically 

 in figures 13 and 14. 



TABLE 33 



The Records of Groups Learning the D Maze after the Partial Learning 



OF the E Maze 



After After After After After complete 



2 trials in E 4 trials 8 trials 16 trials mastery 



Av. Trials. . . . 25.29±2.90 28.43±6.20 26.10±7.46 16.20±7.46 20.86±4.00 

 Av. Errors per 



Trial 2.09 ± .31 1.94± .77 2.64 ± .63 1.07± .40 1.03 ± .35 



Av. Time per 



Trial 38± .08 .32± .06 .31± .03 .41± .17 .91di .25 



TABLE 34 



The Percentages of Transfer of Training in Groups Learning the D Maze 



AFTER the Partial Learning of the E Maze, on the Basis of the 



Average Errors and the Average Time per Trial 



After After After After After complete 



2 trials in E 4 trials 8 trials 16 trials mastery 



Trials —3.73 —15.24 —5.81 34.63 15,44 



Errors —10.60 —2.65 —39.84 43.50 44.97 



Time 75 11 78.76 79.62 72.82 40.13 



40 



