208 



AUDITORY SIGNALS 



on submarines. We will not know the ex- 

 tent of their possible uses until more 

 practical development work has been done. 



Basic Psychological Problems 



There are two ways of approaching the 

 problems of auditory signaling. The first 

 way is to design various sets of signals and 

 try them out. It is a cut-and-try method, 

 and is essentially the way the Flybar signals 

 were developed. We do not know enough 

 about the psychology of hearing to determine 

 beforehand which signals will work and 

 which will not. It is true that we have 

 enough information so that we do not have 

 to start completely from scratch, but a great 

 deal of trial-and-error is still necessary. 

 With a little luck, such a procedure might 

 turn up very useful signals, and might be 

 the fastest way of providing signals for any 

 given kind of information sj'^stem. 



The second approach is to do basic re- 

 search in those problems of hearing which 

 are most relevant to the design of auditory 

 signals. If we had sufficient information 

 about the psychology of hearing, it would 

 be possible to design a set of auditory signals 

 with considerably less guesswork. The cut- 

 and-try procedures could be shortened or 

 even eliminated. Basic research relevant to 

 problems of auditory signaling has the addi- 

 tional advantage of providing data that are 

 not related exclusively to a particular infor- 

 mational system. The data would be useful 

 in the design of auditory signals for any and 

 all purposes. In the long run, it is probable 

 that time spent on basic research would be of 

 more advantage than time spent on cut-and- 

 try designs of specific signalling systems. 



In the rest of this section, we shall discuss 

 some of the basic psychological problems 

 which are relevant to the design and use of 

 auditory signals. For example, suppose 

 that a signal is required which will provide 

 continuous information about one meter 

 reading. If there is a "correct" reading, 

 and action is required whenever the reading 

 becomes too high or too low, then the listener 



must know when the reading becomes too 

 high or low. He must, in other words, be 

 able to discriminate between different tones 

 in the signal. If, in addition, the listener 

 must know how much too high or too low the 

 reading is, then the auditory signal must 

 have some quantitative aspect. Thus we 

 get into problems of psychological scaling, 

 which are problems of the quantitative di- 

 mension of tones. 



Likewise, if we want to transmit lateral 

 directional information — to a radar operator, 

 for example — then we need to know the 

 various ways in which a lateral localization 

 of a tone source can be simulated. We then 

 face the problems of binaural localization 

 of tones. Basic research on this problem 

 will make it possible to provide a lateral 

 indication with little difficulty. In the Fly- 

 bar experiments, as an illustration, a great 

 deal of time was spent in attempting to find 

 a good azimuth indication because it was 

 not known at that time which types of signal 

 will provide lateral displacement of an ap- 

 parent tone source. If more basic informa- 

 tion had been available, this time could have 

 been saved. 



These basic psychological problems, then, 

 are the problems which will arise when an 

 auditory signaling system is designed. They 

 are the problems about which we need much 

 more information if an auditory signal is to 

 be designed with a minimum of guesswork. 



Simple Discrimination 



By simple discrimination we mean the 

 ability to distinguish between two pure tones 

 in either intensity or frequency. It is not 

 often in real life that we are called upon to 

 discriminate between two tones differing only 

 slightly in intensity or frequency, and yet 

 these basic functions provide vital informa- 

 tion by the use of tones. Since, as we 

 pointed out above, absolute psychological 

 quantity is a very nebulous thing, most 

 information must be transmitted by means 

 of differences between tones. The listener 

 must always have a standard tone which 



