Thresholds for seeing visual phantoms and moving gratings.

Abstract

Moving visual phantoms look like real gratings and generate movement aftereffects as do real gratings. Three experiments have been carried out to test whether phantoms resemble real gratings in other ways. Changes in the spatiotemporal frequency of an inducing grating effect the minimum contrast at which it will induce phantoms, but only insofar as the changes affect the visibility of the inducing gratings. Changes in the areal extent of the inducing grating and in the extent of the gap across which phantoms must be induced alter the visibility of the phantoms but, again, only to the extent that the changes affect the visibility of the inducing gratings themselves. Phantoms induced by high-contrast gratings may be cancelled by gratings of near-threshold contrast when the latter are presented 180 degrees out of phase with the inducing gratings. It is concluded that the appearance of phantoms is governed solely by the detectability of their inducing gratings, but that the effects of phantoms on pattern-sensitive channels are not the same as are the effects of real gratings. The pattern characteristics of phantoms thus do not represent a spread of energy from inducing to phantom regions of visual space or their neural representations.