214 



AUDITORY SIGNALS 



problem of stimulus interactions. Hirsh 

 (17), for example, has recently shown that 

 the masked threshold of a tone depends on 

 the interaural phase relations of the binaural 

 tone. Fig. 7 illustrates the amount of this 

 effect for one condition. The use of inter- 

 aural phase differences to simulate localiza- 

 tion, then, would have to take this changed 

 threshold into account. Experiments are 

 needed to determine whether the same kind 

 of a change can occur for the loudness of a 

 tone as well. It is quite likely that both the 

 pitch and the loudness of a tone may also 

 depend on the interaural phase relations. 



30° 60° 90° 120° 150° ISO' 



INTERAURAL PHASE RELATiON OF THE TONE 



Fig. 7. The effect of interaural piiase relations 

 on the binaural masked threshold 

 The ordinate shows the threshold of a 200 cps 

 binaural tone relative to the monaural threshold, 

 at different interaural phase relations. When the 

 ordinate value is negative, the binaural threshold 

 is higher than the monaural threshold, indicating 

 inhibition. Positive values indicate summation. 

 (After Hirsh, 17) 



Another type of interaction has interesting 

 possibilities. Hughes (18) demonstrated 

 that interaural loudness matching is more 

 precise when the two ears are stimulated 180 

 degrees out of phase than when the two ears 

 are stimulated with the same phase relations. 

 Unfortunately, his data are somewhat lim- 

 ited. More research along these lines, how- 

 ever, might prove very fruitful. 



Other Auditory Illusions 



Simulated sound localizations are auditory 

 illusions. The listener hears something 

 which does not really exist, because the 

 stimuli with which he is presented are mis- 

 interpreted. Auditory localizations are not 



the only possible auditory illusions which 

 can be produced, however. Illusions of 

 movement can be produced, as demonstrated 

 in the Flybar experiments. A considerable 

 amount of research is needed, however, on 

 illusions of movement and on any other pos- 

 sible type of illusion. There are practically 

 no leads for research in this area, and yet 

 data here could have great bearing on the 

 future of auditory signals. 



Psychological Scaling 



If auditory signals are to be used to trans- 

 mit quantitative information, i.e., informa- 

 tion about real quantities, then the auditory 

 signal must itself have some quantitative at- 

 tribute. For example, it will not be enough 

 to know that one tone is louder than another; 

 it must also be possible to know how much 

 louder. A psychological scale for loudness 

 has been established by Stevens (30), and one 

 for pitch by Stevens, Vollanann, and New- 

 man (33). These data on pitch and loud- 

 ness scales are adequate for many purposes, 

 but do not provide answers for all the ques- 

 tions. For instance, is the pitch scale the 

 same for tones having a high harmonic con- 

 tent as for pure tones? Are the scales 

 affected by the introduction of a constant ref- 

 erence tone? The following example illus- 

 trates one kind of problem that needs to be 

 solved before quantitative information can be 

 transmitted wdth auditory signals. 



Suppose that a reference tone is followed 

 by a comparison tone, and the intensity of 

 comparison tone with respect to that of the 

 reference tone indicates some quantitative 

 information. Zero quantity would be rep- 

 resented by equality of the two tones. A 

 quantity of five, say, would be represented 

 by plus or minus three db. How much 

 would 10 then be? Is the quantity of plus 

 10 best represented as twice the loudness of 

 the tone that indicates a quantity of plus 

 five, or by twice the difference in loudness 

 between the standard and comparison tones? 

 In loudness units, let us say that the refer- 

 ence tone has a loudness of 1000 millisones. 



