INTEGRATION OF MOVEMENTS IN LEARNING IN THE RAT 485 



produce imperfect responses before the plane. Then, with or 

 without a change in the position of the rat, one or more ineffec- 

 tive thrusts are produced before an effective one plunges the 

 plane. Progression from, and a return to the plane may occur 

 before an effective thrust plunges the plane. These ineffective 

 extensor thrusts occur not only before directive integration is 

 facilitated, but also afterward. On the other hand, when the 

 reflex thrusts are undeveloped, or an unequal antagonistic action 

 of extension and flexion of the limbs exist, fluctuations in the 

 thrusts condition a similar performance, but many advances to 

 the plane usually occur. In addition, it will be noted in other 

 tables that fluctuations in the extensor thrusts will condition 

 the use of another method not commonly used. In the latch- 

 box problem the fluctuations in the thrusts have been noted to 

 occasion the use of the teeth to raise the latch, when previously 

 the neck reflex movement was employed; and in the inclined- 

 plane problem a change in method results from similar fluctua- 

 tions. Usually a change in method in the inclined-plane prob- 

 lem comes about by the performance of an ineffective method 

 commonly used, or after the occurrence of an imperfect response. 

 It has been noted, that when the extensor thrusts are fairly 

 well developed, a change in method does not follow the ineffec- 

 tive performance of the extensor thrust to plunge the plane. 

 Sometimes there appears to be no difference between the produc- 

 tion of an imperfect response and the use of other methods than 

 that of the extensor thrust. 



The substitution of one method for another does not occur 

 with great frequency, but when for instance the modified thrust 

 is used for a time, another method for a trial or two may be 

 used. Often when running and stepping on the plane does not 

 plunge it, or the plane is bolted, progression is reversed, retrac- 

 ing of steps to the plane occurs, and the plane is plunged not 

 from the usual side, but from the other side by the modified 

 extensor thrust. Or, if this last method proves again to be 

 ineffective, the usual running and stepping on the plane in the 

 customary direction is once more resorted to. This change in 

 method is often indicated in tables 13, 14, and 15, when fluctua- 

 tions in the extensor thrusts occurred. 



