(psychophysics) the difference between two stimuli that (under properly controlled experimental conditions) is detected as often as it is undetected