Constant errors in psychophysical procedures are known to increase in magnitude with the intensity of the standard, and with the temporal or spatial separation of the standard and comparison stimuli. It is often assumed that this is due to increasing ‘adaptation’ of the effective standard. It is argued here that the increase is due to a statistical artifact, since the measures of constant error and of differential threshold normally used are not independent, so that any factor which increases the difficulty of discrimination will increase the magnitude of the constant error. It is shown that for differential thresholds ranging from 1 to 25 g. the absolute size of the constant error increases with the size of the threshold, but the direction of the error is variable and remains unexplained.
No takes yet. Share an insight, caveat, or question.
Helen E. Ross (1964) studied this question.