A B S T R A C T S
(C O N F E R E N C E   P R E S E N T A T I O N S)

Recently, Spillmann et al (Investig. Ophthalm. Vis. Sci., 34, 1031, 1993) investigated an intriguing effect of paradoxical motion (PM) seen when a disk filled with counterphase rows of vertical bars is surrounded by a concentric ring equally filled with horizontal bars. The effect is paradoxical in that the disk appears to float relative to the ring but there is no concomitant awareness of displacement. We have discovered four new facts which may help elucidate how ocular fixations interact with spatial and luminance information to yield such PM effects. (1) Retinal adjacency is necessary. In dichoptic presentations (for instance, disk to the left eye and ring to the right eye), PM is not seen whereas it is readily seen monocularly. (2) Retinal inclusion controls which part of the pattern carries the PM effect. Upon adding a second concentric ring with bars having the same orientation as those on the disk, the disk stabilizes and the first ring starts to float. Placing the first ring on a different stereoscopic plane than the other two surfaces does not alter this outcome. (3) A difference in the orientation of the bars is not necessary. A weaker form of PM can be seen in patterns with bars of equal orientation but different contrast or spatial frequency. In these cases, however, the bars appear to drift very slowly and only in one direction. (4) The effects of orientation, contrast, spatial frequency, and stereoscopic depth separation can be combined to create intriguing PM effects.
Keywords: Paradoxical motion, Ouchi illusion, Eye movements

Download (8K) coverage in Pour la Science (French edition of Scientific American)
Visit Nicola Bruno's website 

The present experiment forms part of research on naive physics revealing the 'Aristotelian' tendency of visual perception, ie the fact that subjects describe dynamic events they have seen according to Aristotle's rather than Galileo's physical theory.
In his "De Caelo", Aristotle maintained that once projectiles have been fired they continue to accelerate, and that only in the second part of their trajectory do they decelerate until they stop. This incorrect idea was believed to be true until the time of Descartes (who, in a letter to Mersenne, showed he still believed it), and was taken for granted in Renaissance ballistic manuals.
In our experiment, subjects were presented with computer-generated animations of the trajectories taken by three different projectiles: a cannon-ball, an object sliding on a smooth surface (ice), and an arrow fired from a bow and striking a target. Subjects were asked to change the acceleration of either the first or the second part of the trajectories shown on the computer screen, by choosing various combinations of uniform motion, positive acceleration or negative acceleration.
In virtually all cases, the chosen trajectory of the projectile showed acceleration in the first part, in accordance with the Aristotelian theory.

Paolo Bozzi performed, in the late '50s, what is probably the first experimental work on naive physics. On this topic he has published (in Italian) several papers, and the autobiographical book Naive Physics (Fisica Ingenua. Milano: Garzanti, 1990). See, for a brief description of part of his work,
Naive Physics: An Essay in Ontology
Smith, B., Casati, R.
Philosophical Psychology, 7/2 (1994), 225-244.