116 VINNIE C HICKS \.\l> II. A. CARE 



VI. RELATION OF THE LEARNING CURVES TO INTELLIGENCE 



Our results indicate that the two features of any set of results 

 which serve best as the main criterion or index of the intelligent 

 capacity of a group of animals are the gross values eliminated 

 during learning, and the rapidity with which errors are eliminated 

 relative to surplus distance. 



This conclusion is interesting because it runs diametrically 

 opposite to current theory and practice. There have been in 

 the main two assumptions as to the relation between a learn- 

 ing curve and the intelligent level of a group. 



i. Both comparative and human psychology have tacitly 

 assumed, at least so far as practice is concerned, that the rating 

 of any two groups as to general intelligence is to be based solely 

 and directly upon the number of trials involved in mastering a 

 problem. This assumption is perhaps valid in memorizing non- 

 sense material in such experiments as the Ebbinghaus type, 

 in which the various trials are relatively homogeneous as to 

 the amount of effort expended. In problems of a motor char- 

 acter (and this applies to all animal experiments and to much 

 human work) the various trials are so radically disparate as to 

 the time and effort expended that the assumption is without 

 any rational basis. Moreover, the assumption is controverted 

 by the facts of this paper. However, the inference must not 

 be made that no causal relation obtains between general intel- 

 ligence and the number of trials. We object to relying upon 

 this feature as the sole or main criterion in rating general ability. 



The position may be taken that the number of trials may be 

 regarded as an accurate index of special ability — the ability to 

 learn the maze problem. The writer is frankly skeptical of 

 such an assumption although the point is irrelevant, for com- 

 parative psychology has not been using the criterion in this 

 manner. It has been rating groups of animals as to their order 

 in the intelligent scale, and this means general and not special 

 ability. 



2. Comparative psychology has also broached the proposition 

 that the rational status of a group of animals can be legitimately 

 inferred from such a feature as the " slope " of a learning curve. 

 This assumption does not refer to general intelligent ability, 

 but to rational ability. 



Thorndike assumed that a gradual slope of a time curve is 



