Goldstein E B, Fink S I
J Exp Psychol Hum Percept Perform. 1981 Oct;7(5):954-67. doi: 10.1037//0096-1523.7.5.954.
These experiments show that observers can selectively attend to one of two stationary superimposed pictures. If superimposed line drawings are presented to observers who are told to attend to one line drawing in the pair and to ignore the other line drawing in the pair, then a subsequent recognition test in which the pictures are presently singly, the attended picture in each pair is recognized much more frequently than the unattended picture in each pair. This selective recognition occurs both with large (11 degrees-22 degrees) displays in which observers are free to make eye movements during a 3-sec exposure and with small (1 degree) displays in which observers are instructed to fixate steadily on a point during a 1-sec exposure. The results of the steady fixation experiments show that in the absence of eye movements, attention to one of two superimposed stimuli can cause an observer to remember the attended image and not to remember the other, clearly visible, unattended image in a superimposed pair.
