The effect of adaptation on subjective brightness

Abstract

Attempts have often been made to find a relation between stimulus and sensation magnitudes since the implications of the existence of absolute and differential threshold, and theories of intensity discrimination, flicker sensitivity, resolving power, and so forth must depend largely on the view taken of this relation. Some of the problems have been discussed by wright (1938). Two principal methods have been employed in the case of vision: first, the direct estimation of sensation intensities as numerical multiples of one another (Richardson 1929; Maxwell 1929;) Hopkinson 1939) or as intermediate between other sensation intensities (Gage 1934); and secondly, the integration of theI/∆Icurve, on the assumption that just noticeable differences at all points of the sensation-intensity range represent arithmetically equal sensation steps (Fechner 1858; Abribat 1935). Both these methods have been criticized on the general ground that sensation intensity is not a measurable quantity, and that the numerical results obtained are therefore illusory. In the case of the first type of experiment it is held that the numbers are merely names applied by convention to sensory intensities which can only be ranked in an ordinal series; and in the case of integration of∆I’s it is objected that there is no method of showing that j. n. d.’s at different sensation levels represent equal increments in sensation intensity. It has also been shown by Gage that even if the assumption of the equality of j. n. d.’s were correct the result obtained would be false, since the effect of adaptation in altering the sensation intensity after initial exposure to a changed stimulus is neglected.