Derivation Of Brightness Scales Using Partition Scaling

To capture the relationship between the magnitude of an optical stimulus (e.g. luminance) and its perceived brightness, a number of psychophysical procedures can be used. One of the oldest procedures capable of capturing such a relationship is partition scaling (PS). A PS procedure constructs an interval scale of a psychological attribute (e.g. brightness) directly from the judgements of observers. It is rarely used in visual research, despite being one of the oldest psychophysical procedures. PS is investigated on its robustness, susceptibility to possible biases and its speed. Partition scaling experiments with simple achromatic discs were set up to obtain brightness scales as a function of luminance. In total four luminance ranges were investigated: three subranges (low-, mid- and high-range) equally divided from 5 to 175?cd/m(2) and one overlapping full-range. Results show that observers had difficulties in accurately estimating the mid- and high-ranges, because the brightness of the two anchor discs was too similar. However, perceptual brightness scales could be obtained for the full- and low-range. To reduce cumulative errors and increase observer accuracy, an improved PS method was implemented. Results show that PS is a valid, rapid psychophysical procedure capable of capturing a brightness scale without the need of prior knowledge of the psychophysical brightness function.