SUMMARIES AND CORRELATIONS. 251 



While they do not affect the general tendency, they do save generali- 

 zations from the error of artificial simplicity. We are consequently 

 under a double obligation to examine in some detail the apparent 

 exception to the main tendency of our results. 



In discussing our eye-reaction technique, we found some grounds 

 for dissatisfaction owing to the limited number of positions for the 

 peripheral object of regard, and the consequent possibility of antici- 

 patory reactions. The same fault will be found (p. 89) to have pro- 

 duced an unexpected practice effect in the eye-reactions on normal 

 days. We can not agree with a supposititious critic who, on the ground 

 of this practice effect, might hold that the eye-reaction fails to fulfill 

 our demands for a thoroughly practiced experimental process. That 

 which is thoroughly practiced in this reaction is, however, the differ- 

 ential coordination of the eye-muscles to bring the line of regard to any 

 one of an indefinite number of positions. Our experiment was an arti- 

 ficial simplification of natural conditions. Instead of an indefinite 

 number of possible positions we used only 6. Apparently all our 

 subjects learned by experience during the experiments to respond to 

 one of the 6 new positions more rapidly than they were in the habit of 

 responding to an indefinite number. Doubtless this should have been 

 foreseen in planning the experiment. In excuse one can only say that 

 the data on normal eye-movements are not very abundant and the 

 particular point had never arisen before. Dodge 1 had found that in 

 the course of over 10 years of eye-reaction records his eye-reaction had 

 not materially changed and we failed to realize that in his experiments 

 a great variety of positions were used. It is not impossible that indefi- 

 nite variation of the eye-reactions would have been open to more serious 

 criticism because of lack of uniformity on the different experimental 

 days. After all, as far as the main results are concerned, a moderate 

 practice effect is not serious. It was provided for by the distribution 

 of normal days. This type of reaction gave comparable values for all 

 sorts of untrained subjects, and the effect of repetition is clearly repre- 

 sented on the normal base-line. 



Our facilitation-inhibition problem, however, gives the possibility of 

 simplified elaboration of reaction a more serious aspect. We may 

 indicate its bearing by a question: "What would have happened if we 

 had still further simplified the motor elaboration of the eye-reaction by 

 reducing the number of stimulus positions to one instead of six?" The 

 answer to this question we know from accidental experience. Such 

 simplification would have led to frequent if not to regular anticipatory 

 reactions. The voluntary control of our eye-movements is meager at 

 best. If we know where an object is about to appear it takes a great 

 deal of practice and an entirely artificial inhibition to prevent looking 

 at the expected place. The artificial development of such inhibitions 



1 Dodge, Monograph Supplement of the Psychol. Review, 1907, No. 35. 



