AUDITORY PATTERNS OF THE AURAL MEMBRANE 307 



gradient pattern on the aural membrane. By a method of localized 

 measurements along the cochlea, Culler [1935] and later Kemp and John- 

 son [1939] were able to map the various maximal positions of pitch 

 responsiveness in the cochlea in terms of distances from the apical end. 

 Figure VII-25 shows such an auditory map, in which the spiral repre- 

 sents the cochlea to scale with its length divided into 100 units. 



Further evidence to support this map of positions of maximum 

 response to pure tones is offered by Fletcher [1940], who carried out 

 two sets of calculations, the first based on the width of the band of 

 frequencies that a noise must contain in order to mask a pure tone, and 

 the second upon the ratio of the intensity of the necessary noise to the 

 intensity of the masked tones. 



On the basis of these independent supporting data we can now accept 

 with confidence the pitch maps of the cochlea. In Fig. VII-25 the height 

 of the dark portion is drawn proportional to the response of the nerves 

 at the positions indicated. The pattern is that produced in a typical 

 normal cochlea when a 700-cycle tone is heard at an intensity level 

 90 db above threshold. Most of the experienced loudness is seen to 

 come from position 20. A considerable part of the loudness is shown to 

 originate near positions 36, 48, and 56. These last are the aural har- 

 monics probably introduced by the asymmetrical characteristics of the 

 ear at this loudness level. The total loudness is represented as corre- 

 sponding with the total area of the darkened part. As the intensity 

 increases the amplitude of the loudness pattern increases, spreads, and 

 additional higher harmonics appear further down the aural membrane. 



An important application of pitch localization, as used by Fletcher, 

 is illustrated in an audiogram of a case of nerve deafness (Fig. VII-26). 

 The breadth of the cross-hatched portion is supposed to indicate the 

 amount of damage to the nerve endings at the various positions along the 

 basilar membrane. In order to hear any sound, the deaf person must 

 have the normal auditory pattern (solid black) projecting above his 

 nerve-deafness pattern. The 6900- and 4600-cycle tones will not be 

 heard, for their auditory pattern does not project through the deafness 

 pattern. 



The 2300-cycle tone, however, will be heard only in proportion to 

 the amount of the loudness pattern projecting above the deafness pattern 

 at that frequency. 



The above evidence strongly favors a " place-resonance " theory of 

 pitch perception, in which loudness correlates best with the total num- 

 ber of active nerve fibers rather than with any simple function of the 

 number or frequency of nerve impulses. 



The above general principles are basic to any understanding of the 



