BASIC PSYCHOLOGICAL PROBLEMS 



213 



and Ford (8) have shown that the normal 

 person's abihty to equate the intensities of 

 tones in the two ears is not nearly as good 

 as would be predicted from the data on 

 localization. A much greater intensity dis- 

 parity between the two ears is needed in 

 order to produce an apparent displacement 

 than seems reasonable. In addition, the ex- 

 periments by Garner showed that many ob- 

 servers have two ears that do not agree about 

 the loudness of a given tone intensity. For 

 example, an intensity of 60 db in one ear 

 may be equally loud to an intensity of 70 db 

 in the other ear. The ears are, in a sense, 

 off balance in loudness when both ears are 

 stimulated with the same intensity. This 

 same effect had also been shown incidentally 

 by Forbes et al. (7). 



Phase Differences. Phase differences at 

 low frequencies have shown a somewhat bet- 

 ter picture in some respects. The most re- 

 cent study of threshold of detection for phase 

 differences was reported by Hughes (19). 

 For experienced subjects, the threshold for 

 the detection of a shift of the tone from the 

 median plane was about 20 degrees phase 

 difference at the lower frequencies. This 

 figure was closer to 40 degrees for inexperi- 

 enced observers, however. Nevertheless, 

 these figures agree with the Stevens and 

 Newman data much better than the data 

 for intensity differences do. But even 

 Hughes' data indicate how unreliable is such 

 a simulated localization. For example, his 

 experienced subjects detected an abrupt 

 change from zero degrees phase difference 

 to 90 degrees phase difference only 80 percent 

 of the time, at the lower frequencies. There 

 seems little question that attempts to simu- 

 late apparent localization with the use of 

 simple phase or intensity differences between 

 the ears will not produce the positive stimu- 

 lus required. 



Comhination Cues. The final answer to 

 the localization problem, on which more re- 

 search is definitely needed, may be that 

 good simulated localization can be produced 

 only by the combination of two or more of 



the cues. In an extensive series of experi- 

 ments, Langmuir et al. (22) simulated a 

 sound source by practically duplicating the 

 interaural differences which occur in real 

 situations. The task of their observers was 

 to make the sound appear to be in the me- 

 dian plane, which they did by adjusting the 

 apparatus. Thus they heard the sound ap- 

 pear to change location, and eventually "cen- 

 tered" it. Their results show an ability to 

 localize the sound source in the median plane 

 with a precision greater than that shown in 

 the Stevens and Newman experiments. 



Dynamic Cues. This greater precision of 

 localization suggests that dynamic, or mov- 

 ing, cues are much better than static cues. 

 In the Stevens and Newman experiment, 

 the observers made a static localization; i.e., 

 they determined the azimuth location of a 

 tone source which was stationary. In the 

 Langmuir et al. experiment, the subjects were 

 able to listen to all simulated positions, and 

 hear the changes as they manipulated the 

 apparatus. Thus they had many of the 

 dynamic cues which are available in real 

 situations — cues which depend on movement 

 of the sound source or movement of the head 

 and body with respect to the source. Wal- 

 lach (36), too, has emphasized the dynamic 

 nature of sound localization in real situa- 

 tions. 



Whatever the final outcome of this type 

 of research, it is clear that more research is 

 needed before it will be possible to simulate 

 good localization of sounds. It may be that 

 the best simulation, or the best artificial 

 localization, will occur with unrealistic de- 

 vices. Langmuir and his associates, for ex- 

 ample, found that localization was very good 

 when the observers could adjust the sounds 

 to give a binaural null at a position corre- 

 sponding to straight ahead. This method 

 does not provide a realistic apparent sound 

 source, but does provide accurate positioning 

 on the part of the operator. 



Stimulus Interactions: Phase Effects. The 

 sound localization experiments will in them- 

 selves make more research necessary on the 



