n inaur.il i.oc.rriON oi- complex sounds 41 



the effect is at first quite alarming — one seems to be in the middle 

 of a pack of dogs some of which are rushing viciously at one's throat." 

 An illustration of failure to form any image is found in a phenomenon 

 observed in the use of binaural compensators for determining the 

 direction of submarine sounds. The sound is picked up by two sub- 

 marine telephone transmitters and led to the ears through inde- 

 pendent paths. By adjusting the lengths of the paths the image 

 can be shifted from side to side and for practical purposes the setting 

 of the instrument is made by bringing the image exactly to the middle. 

 A fairly definite sound image is formed, but observers report that part 

 of the sound does not merge into this sound image and move in re- 

 sponse to the adjustment, but instead appears as a diffuse back- 

 ground of noise.® This may be explained on the assumption that, 

 while the images formed from most of the sound components agree 

 sufificiently well that the observer corrects them to a single position, 

 certain components are so distorted by resonance effects inherent 

 in the apparatus that their images are scattered more or less at ran- 

 dom. The lack of agreement among any considerable number of 

 these prevents the formation of a second image and causes the sense of 

 diffusedness. 



As the distortion becomes still more extreme we should expect 

 the experimental results to depend more and more upon the observ- 

 er's power of resolution, for as the distortion is progressively in- 

 creased a condition must finally be reached where the positions of the 

 images are appreciably different for two components whose frequencies 

 are so nearly alike as to make their recognition as separate tones 

 dillficult if not impossible. This condition actually occurred in an 

 experiment of Baley's with a sound consisting of a mixture of sus- 

 tained tones. Its effect on the listener is interesting from the stand- 

 point of subconscious readjustment of discordant data. 



Baley's ^ experiment consisted in applying a number of sustained 

 tones to one ear of a musically trained observer and a number of differ- 

 ent tones to the other ear, and testing his ability to assign them to 

 their proper sides. So long as the intervals between the tones were 

 fairly large, the observer never failed to locate them correctly. Con- 

 sidering the entire stimulus as a complex sound we may think of the 

 observer as locating the tones individually and finding them to fall 

 definitely into two groups whose images are located one at each ear. 



* This interesting phenomenon was called to our attention by Mr. Richard D. 

 Fay of the Submarine Signalling Corporation who tells us that it has been noted 

 by a large number of observers. 



■Stephan Baley: Zeit. f. Psycho!, u. Physiol., v. 70, 1914, p. 347. 



