26 THOMAS WILLIAM BROCKBANK 



As has been shown in the foregoing pages, in the process 

 of learning, in the acquirement of a habit, the establish- 

 ment of complex integrations of movement has a very 

 marked effect on the retention of the habit. All methods 

 of learning produce finally the same result, viz, the estab- 

 lishment of the perfectly integrated habit. Yet since each 

 method of learning, according to the distribution of effort, 

 has such a marked effect on the characteristics and general 

 progress of the learning process, a legitimate working 

 hypothesis may be laid down to the effect that retention 

 and subsequent redintegration may be affected in a some- 

 what similar manner by the method of distribution of 

 trials previously employed in learning. Within a definite 

 limit it is the object of the present chapter to show some 

 effects of the distribution of effort in learning on the re- 

 tention and redintegration. 



It was with the twofold object of attempting to throw 

 some light on retention that the different groups of rats 

 considered in the foregoing pages were divided into three 

 trials per day groups and one trial per day groups. The 

 " error " Tables I-A, II-A, III-A, and IV-A have already 

 been explained in reference to the dominant " error." The 

 economy of these same tables is summed in the " error " 

 totals. It may be noted in these totals that the number 

 of " errors " in learning were far more numerous for the 

 three trials per day group than for the one trial per day 

 group. This fact is significant in the explanation of the 

 economical superiority of the one trial per day method 

 over the three trials per day method. The group following 

 the three trials per day method display a more pronounced 

 difficulty in establishing the perfect integrations of move- 

 ment in the solution of the successive turns and the con- 

 sequent acquirement of the habit, if one is to judge from 

 the " errors." The three trials per day method is pro- 

 ductive of the greater number of " errors," and evidently 

 a greater expense of energy. Therefore, the one trial per 

 day method is the method of wider distribution of trials 

 and is more economical. 



Turning to the influence of retention, in redintegration, 

 it may be seen that a somewhat similar situation exists. 



