Abstract
Estimates of the subjective magnitudes of flicker sensations correlate poorly with the amplitudes of electrical brain responses (evoked potentials). A power function of exponent between 0·9 and 1·5 describes the relation between stimulus and subjective sensory magnitude. On the other hand, the relation between stimulus and objective measure varies with electrode position, is different for different frequency components of the evoked potential, and may not be a power function at all. As a rough generalisation, correlations between evoked potentials and perception are poorer when spatially-unpatterned stimuli are used than when the stimulus is spatially patterned.