REFLEX AND CONDITIONED ACTION 6^ 



but by countless others. The engram of the particular problem- 

 box habit in question is certainly not in the precentral gyrus 

 (which alone was injured); it must lie in some premotor arrange- 

 ment of associational neurons not at all affected by the opera- 

 tions. The other cases cited are similarly inconclusive. 



In another place, Lashley says, "it is characteristic that the 

 conditioned reflex is not built up gradually, but is fully integrated 

 when it first appears." Of course, the last statement must be 

 true, for this type of learning does not involve the gradual 

 formation of new elementary behavior patterns but only the 

 doing in unfamiliar situations of perfectly familiar things for 

 which definite neural mechanisms are already laid down in innate 

 or habitual organization. When an auditory stimulus which does 

 not ordinarily produce salivation is diverted to the saUvary path, 

 a fully integrated salivation follows as soon as the proper central 

 connection is made for the salivatory mechanism is preformed 

 and ready to operate perfectly as soon as it is activated from any 

 source. The physiological problem is. How is the central transfer 

 made from the "natural" auditory efferent path to the "natural" 

 gustatory efferent path leading to the saHvatory nucleus? The 

 evidence is clear that this reorganization of the conduction 

 pathways within the correlation center is usually very grad- 

 ually effected. It may require a hundred experiences to per- 

 fect it. 



This is the way learning is usually effected in lower animals. 

 But in higher mammals, where cortical associational tissue is 

 abundantly developed, the picture is different. In man and the 

 larger apes (Kohler, 1925) a new combination of the elements of 

 a behavior pattern or an adaptation of familiar behavior patterns 

 to a new situation may appear suddenly when the animal "sees 

 through" a problem of conduct to an immediate solution without 

 overt trials or random fumbling. This type of behavior has not 

 been discussed in this chapter. It involves the use of a sort of 

 neural organization quite different from that which is adequate 



