SECTIONAL TRANSACTIONS— A. 301 



problem for the psychologist, but if its solution is to be attempted quantita- 

 tively that must rest on the quantitative systems and methods developed by 

 the physicist as the only practicable basis. The study of the differences 

 betvi^een individuals in sensory experience is a second psychological problem, 

 and this too, if its solution is to be attempted in quantitative terms, depends 

 no less on the quantitative relations and concepts developed by the physicist, 

 and most emphatically on the physicist's units of measurement. Apart 

 from quantitative relations, concepts, and units, developed from a point of 

 view which eliminates the individual observer, no solution in quantitative 

 terms of either of these psychological problems is possible. There are not 

 two worlds of sense, one in the mind studied by the psychologist, and 

 another outside studied by the physicist. There is one world of sense 

 studied by both psychologist and physicist, and the units and methods of 

 measurement from the nature of the case must be those of the physicist. 



Dr. J. H. Shaxby. 



The loudnesses of sounds, though capable of being specified as possessing 

 magnitude, yet cannot be arranged on any single definite and unequivocal 

 scale. The method of measurement affects the magnitudes assigned to a 

 given series of sounds. In particular, phenomena such as sensory adaptation 

 make the formulation of a scale of sensations a dubious procedure. 



It is, further, doubtful whether any estimate, other than the purely intro- 

 spective, really gives a measure of loudness so much as of the physical 

 intensity associated with sounds arranged on a scale more or less arbitrarily 

 assumed to be one of loudness, or with the physiological processes of the 

 auditory nerve. 



The decibel, though exceedingly useful in specifying intensities, seems 

 to agree neither with a scale of equal increments of loudness as subjectively 

 judged, nor with the successive steps of a scale based on differential 

 thresholds. Even if we concede its use as a ' loudness ' unit, its utility in 

 that respect rests on its supposed parallelism with the scale based on the 

 Weber-Fechner law ; it is more than doubtful whether this law is valid 

 for audition, and in any case the decibel is commonly used between limits 

 far surpassing the possible range of Weber's law. 



Dr. Wm. Brown. 



Although G. T. Fechner's assumptions as regards the measurement of 

 sensation intensities are not theoretically justified, the possibility of direct 

 mental measurement in terms of contrastes sensibles or ' sense-distances ' 

 was conclusively demonstrated by J. R. L. Delboeuf as far back as 1878. 

 The Weber-Fechner law can thus be rewritten in the form 



SSo = ^ log p- 



where SSo represents a sense-distance, and Sq is any finite intensity of 

 sensation taken as the conventional zero (not necessarily liminal) ; R, Rq 

 are corresponding stimulus values. 



At the two extreme ends of the scale there are deviations from the 

 logarithmic law of a continuous and uniform nature. 



I'he method of constant stimuli, which is based upon the accumulation 

 of large numbers of observations and the employment of statistical methods 

 in drawing conclusions from them, is the most reliable psycho-physical 

 method for solving the quantitative problems of psycho-physics. 



