Experiments have verified the observations of Corbett and White that people are more sensitive to flicker on raster-scanned displays (e.g., television and visual display terminals) if the field scan direction is changed from the conventional top-to-bottom direction to bottom to top, left to right, or right to left. When measured by means of the flicker fusion frequency (FFF), this effect is present only when normal eye movements are permitted. With fixation, the FFF is similar for all scan directions. The effect is demonstrated by the different frequencies at which a directional component is detected in the flicker (which we have called the rolling flicker threshold frequency, or RFTF); the RFTF is lower for top-to-bottom scanning. The effect does not appear to be associated with any asymmetry in eye movements or with eccentricity of the retinal image. The hypothesis that this difference is a consequence of adaptation to televisions and VDTs is supported by the observation that the effect's magnitude correlates with daily exposure time to televisions and VDTs and by a pilot study showing that the effect is reduced when one views a television scanned from bottom to top for 1 h.