LOUDNESS 



423 



of observations by the rest of the crew, (a total of eleven observers) are 

 shown in graphical form in Fig. 17. The arrow indicates the median 



^ 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 



PER CENT 



Pig 17 — Percent of observations estimating 1000-cycle tone to be louder than 125- 



cycle tone. 



level at which the 1000-cycle reference, in the opinion of this group of 

 observers, sounded equally loud to the 125-cycle tone. 



The testing method adopted was influenced by efforts to minimize 

 fatigue effects, both mental and physical. Mental fatigue and 

 probable changes in the attitude of an observer during the progress of 

 a long series of tests were detected by keeping a record of the spread of 

 each observer's results. As long as the spread was normal it was 

 assumed that the fatigue, if present, was small. The tests were con- 

 ducted on a time schedule which limited the observers to five minutes 

 of continuous testing, during which time approximately fifteen obser- 

 vations were made. The maximum number of observations permitted 

 in one day was 150. 



To avoid fatiguing the ear the sounds to which the observers listened 

 were of short duration and in the sequence illustrated on Fig. 18. The 



WARNING 

 BUZZ 



SOUND 



1000 

 CRS. 



SOUND 



1000 

 CRS. 



0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7,5 



TIME IN SECONDS 



Fig, 18 — Time sequence for loudness comparisons. 



duration time of each sound had to be long enough to attain full loud- 

 ness and yet not sufficiently long to fatigue the ear. The reference 

 tone followed the x sound at a time interval short enough to permit a 



