The effect of voluntary attention on afterimage fragmentation was explored in two experiments. The afterimage, in the form of a 30 degrees-tilted star of David, was generated after prolonged steady fixation in the first experiment, and with a brief and intense flash in the second experiment. Subjects were instructed to select various target shapes in the afterimage for attention and, at the same time, observe what was visible or invisible. Verbal reports and manual responses to afterimage changes were analyzed. Attended shapes were found to disappear from awareness Faster than unattended ones (experiment 1), and complementary shapes were found to predominate visual awareness when one of the pair was selected for attention (experiment 2). Voluntary attention was also found to affect closure (filling-in of enclosed regions) and smoothing of line figures in afterimages.

The apparent size of an afterimage viewed from distances between 5 cm and 580 cm was matched to that of a size-adjustable stimulus at a fixed distance (20, 30, 90, and 200 cm). The experiment was conducted under normal indoor illumination with a procedure that facilitated matching for angular size. The matched size was found to increase with focal distance within 1 m and very little beyond 1 m. Similar results were obtained with an equivalent series of real stimuli subtending a constant visual angle. These findings suggest a scaling in perceived angular size in proportion to the oculomotor adjustments for accommodation and convergence. The observations of perceived angular size of the afterimage complement what Emmert's law is meant to describe (perceived object size of the afterimage), even though as the focal distance decreases it may be increasingly difficult to tease out perceived object size and perceived angular size with the matching procedure.

Whether the Troxler effect (TE) has to do with interocular suppression and/or summation was studied with dichoptically matched (binocular or dioptic) and unmatched (monocular) stimulus presentations. Perceptual disappearance was found to occur more slowly under the binocular condition (mean=14.2s) than the monocular condition (mean=8.4s), but much faster than predicted by probability summation of the experimentally obtained latencies and durations of the TE in the monocular conditions (>27 s), suggesting a binocular inhibitory summation, the opposite of the binocular summation found with detection and contrast matching tasks [(Blake, R., & Fox, R. (1973). The psychological inquiry into binocular summation. Perception and Psychophysics, 14, 161-185; Blake, R., Sloane, M., & Fox, R. (1981). Further developments in binocular summation. Perception &Psychophysics 30, 266-276.)]. In addition, Ss with poorer stereoacuity took longer to see the disappearance in the monocular condition, and showed a larger disparity between the TEs from the two monocular conditions, suggesting a contribution of interocular suppression to the TE.