INTEGRATION OF MOVEMENTS IN LEARNING IN THE RAT 495 



and low points are due to more efficient movements. Upslopes, 

 with increases in time before the problem is solved, represent 

 advances to the plane with wandering progression, downslopes, 

 with decreases in tune, show the occurrence of fewer of these 

 movements, and a greater effectiveness in the method employed. 

 Effective responses produced 2 perfect records. Occasionally 

 fluctuations in reflex excitability decrease activity and increase 

 altitudinal points. When reflex excitability is constant, gen- 

 erally high altitudinal points are not great. Greater altitudinal 

 points, representing the presence of hyponormal reflex excit- 

 ability, and increases in time, are presented in curve IV. This 

 curve shows no great differences from the other 3 curves but 

 reveals more pronounced changes in the organism when learn- 

 ing. From the twenty-third to the thirty-third trial perfect 

 records were made, but none thereafter. The modified extensor 

 thrust was used. 



These curves represent the degree of effectiveness of every 

 movement made by a rat, and this effectiveness is dependent 

 upon the degree of the development of interaction permissible, 

 when, in two rats the functional condition of one organism is 

 fairly, and when in two, poorly developed. These curves show 

 a gradual change in the physiological condition of the organism. 

 Their interpretation rests upon physiological changes, not upon 

 a mathematical calculation of any kind. 



By comparison of the results obtained when investigating 

 learning in the latch-box and in the inclined-plane problem, it is 

 evident that the rat's organized mechanism of reflexes does not 

 work the same in two dissimilar situations. Accordingly, there 

 appears a diversity in the results obtained from the two prob- 

 lems. The situation presented by the latch-box problem is 

 without question best suited for the rat's reflex mechanism, but 

 even here there are limitations, for when rats manifest hypo- 

 normal reflex excitability, posturing to produce the neck reflex 

 is at all times impossible. On the contrary, posturing is not 

 necessary in the inclined-plane problem to plunge the plane. 

 But, proportionally greater is learning in the latch-box, than in 

 the inclined-plane problem. Only in two cases did the latch- 



