INTEGRATION OF MOVEMENTS IN LEARNING IN THE RAT 381 



cessful" movements become, then, of some sensory consequence, 

 instead of some pleasurable consequence, to the organism. 

 Rather the opposite view is expressed by Watson. In his mech- 

 anistic conception of learning, the recency and the frequency of 

 the performance of these movements alone produce the fixation, 

 the retention, of them. The " successful' 7 movements are definite 

 movements which must be integrated to produce a unitary act. 

 Sensory excitations, by selecting neural arcs, produce a " tying 

 together or integration" of these movements. It is obvious that 

 Watson does not wish to imply that these movements may pro- 

 duce any increase in sensory effect, and thereby weaken his own 

 mechanistic explanation of the term repetition ; and Carr, in his 

 belief in an increase in specific sensory intensity, leaves open for 

 suggestion the appearance of a "need." Since the pleasurable 

 tendency is disregarded, the "need," the "psychic," comes into 

 existence for the production of the "successful" movements at 

 the moment the first specific stimulus is received. 



In both these theories, in whatever manner the "successful" 

 movements are retained, the "unsuccessful" are eliminated, dis- 

 posed of, because they have no sensory consequence, or they are 

 not repeated. Then in two different ways the "unsuccessful" 

 movements may be eliminated. For Carr it is the intensity of 

 the sensory impressions brought about by the recency and the 

 frequency of the "successful" act that results in the establish- 

 ment of the permancy of this act and eliminates the "unsuccess- 

 ful" movement. For Watson it is the recency of the performance 

 of the "successful" act with its supposed immediate physiological 

 effect, coupled with the frequency or repetition and its cumula- 

 tive physiological effect by which certain neural arcs become op- 

 erative in producing this act, fixating it, that eliminates "unsuc- 

 cessful" movements. One sort of movements is set over against 

 the other, so that the "successful" movements in one case are of 

 some sensory consequence to the organism, and in the other of no 

 sensory consequence whatsoever, but of some physiological im- 

 portance, and all the other movements are of indefinite value. 



As in previous theories of the kind on learning, the burden of 

 retention of movements is thrown on the nervous system. The 



