DYNAMOGENIC INFLUENCE OF .LIGHT 



361 



tabulation. This measurement establishes the time consumed 

 in making an error as the measure of the penalty imposed. The 

 penalty is arbitrary. In a factory, for example, the penalty for 

 errors would depend on the relative value of the raw material 

 with respect to the finished product; the value of the spoiled 

 material as salvage; the time the machinery must be stopped to 





0, 



t 



-20 



-15 



-iO 



2.09 5 

 Va I ue 5 of 



20 



23 



FIG. 1. PERCENTAGE DISTRIBUTION OF PERCENTAGE DIFFERENCES BETWEEN 

 DAILY PERFORMANCE IN LIGHT AND IN DARKNESS, FOR ALL THE 



SUBJECTS (SEE TABLE 3) 



Ordinates: percentage of measurements included in average. Abscissae: 

 values of A. The difference (A) is expressed as a percentage of the performance 

 in darkness, or 



100 (Pi-Pp) 



The empirical measurements are indicated by crosses. The circles indicate a 

 "smoothed" distribution of the original measurements, obtained by one appli- 

 cation of the "moving average" in groups of five. The curve represents the 

 most probable distribution of a large number of similar measurements about 

 the mean according to the Gaussian law. 



correct errors, etc. In certain circumstances it would be profit- 

 able to increase speed at the expense of accuracy within certain 

 limits; and these limits would be determined by such factors as 

 I have mentioned and would vary in different shops. In this 



P8TCHOBIOLOGT, VOL. II, NO. 4 



