Velocity Constraints on Apparent Rotational Movement

Abstract

Abstract. The visual illusion of apparent rigid rotation was produced by sequential alternation of two views of the same object in different orientations. The minimum stimulus-onset asynchrony required for the appearance of rigid rotation was a linearly increasing function of the angular difference in orientation between the two views. Variation in the size of the object affected the zero-intercept of the function, but the slope was virtually constant. The slope invariance suggests that the appearance of rigid rotation is constrained by an upper bound on the apparent angular velocity of the object as a whole, rather than a bound on the linear velocity of its parts.

Object-based apparent motion

“…According to Cavanagh and Mather, similar comparator mechanisms underlie both types of motion perception. The observation in Experiment 1, which indicated that apparent motion between complementary images yields a relationship between SOA and angular deviation that is the same as that found for intact or complete images (e.g., Farrell et al, 1982;Shepard & Judd, 1976), is consistent with this view. The results also suggest that image-based and object-based apparent motion share the same mode of operation.…”

“…Under suitable conditions, alternating static images that are related by a shape-preserving similarity transformation produce the perception of a rigid motion over a specific trajectory. A minimum temporal separation between the two images is required for such motion, and this minimum increases linearly as a function of the spatial separation between them (Bundesen, Larsen, & Farrell, 1983;Farrell, 1983;Farrell, Larsen, & Bundesen, 1982;Farrell & Shepard, 1981;Korte, 1915;Shepard & Judd, 1976). It is generally assumed that the intercept of this function measures the time it takes to encode the two successive stimuli as different views of the same object, and that the slope reflects the maximum speed of interpolating a path between them (McBeath & Shepard, 1989;Shepard, 1984).…”

“…A rotational transformation was deliberately chosen in the present study, because it has been found to be particularly sensitive to the figural properties of the stimulus (e.g., Farrell et al, 1982;Farrell & Shepard, 1981;Shepard & Judd, 1976). Two successively presented shapes can give rise to the phenomenal experience of a single object in motion when they are related by a shapepreserving transformation (Warren, 1977; but see Pittenger & Shaw, 1977).…”

Apparent motion between shapes differing in location and orientation: A window technique for estimating path curvature

“…A display for which kinematic geometry prescribes a more curved, longer path of transformation should therefore require a longer SOA and hence (for a fixed stimulus duration) a longer lSI. If SOA is too short for completion of the circular path, the perceptual system may connect the alternately displayed stimuli through a path that is flattened and hence shortened enough to allow motion to be completed within the time available (as suggested, also, by the results of Farrell et al, 1982;Proffitt et al, 1988).…”

“…Presumably, this is because more time is needed for the impletion of a longer path. Indeed, the increase tends to be linear, whether the motion itself is rectilinear (Corbin, 1942;see Shepard, 1984, p. 421) or rotational (Bundesen et al, 1983;Farrell, 1983;Farrell, Larsen, & Bundesen, 1982;Farrell & Shepard, 1981;Shepard & Judd, 1916). Moreover, Shepard and Zare (1983), using a new technique of "pathguided" apparent motion, demonstrated that even when the straight distance between two alternately presented dots was fixed, the critical SOA for apparent motion over an experimenter-determined curved path increased linearly with the length of that curved path.…”