Temporal integration in the perception and discrimination of solid shape

Abstract

Two experiments evaluated the importance of temporal integration for the perception and discrimination of solid object shape. In Experiment 1, observers anorthoscopically viewed moving or stationary cast shadows of naturally shaped solid objects (bell peppers, Capsicum annuum) through narrow (4-mm wide) slits. At any given moment, observers could only see a very small portion of the overall object shape (generally less than 10%). The results showed that the observers' discrimination performance for the moving cast shadows was much higher than that obtained for the stationary shadows, demonstrating the ability to temporally integrate the piecemeal momentary information about shape that was available through the narrow apertures. In a second experiment, estimates of the strength of the observers' impressions of solid shapes rotating in depth were obtained as well as discrimination accuracies; perceptions of the original moving condition were compared with a new condition where the frames of the apparent motion sequences depicting solid objects in continuous motion (behind the slits) were randomly scrambled. The observers perceived the anorthoscopic displays as depicting solid objects rotating in depth, but only in the continuous motion condition. Interestingly, the discrimination performance in the scrambled condition remained relatively high-observers were still able to integrate information across the multiple scrambled frames in order to produce discrimination performance that was significantly higher than that obtained in the stationary shadow condition. This study was the first to thoroughly evaluate whether and to what extent human observers can effectively discriminate and perceive solid object shape anorthoscopically.

Aging and temporal integration in the visual perception of object shape

“…Zöllner’s observation was important, because it demonstrated that the human visual system is capable of integrating (over time) the momentarily visible fragments into a complete perception of whole objects. Later investigations have confirmed Zöllner’s observations 1 , 2 , 4 – 7 . As an example, consider Fig.…”

“…We can easily recognize the cat, deer, and parrot despite the fact that significant proportions of all three animals are not visible, because of intervening surfaces, leaves, branches, and flowers. In cases when objects move behind occluding surfaces, human observers can successfully recognize objects even when they are more than ninety percent occluded 1 , 2 . This ability was first studied more than 160 years ago by Zöllner 3 ; he described a phenomenon that he referred to as “anorthoscopic” perception.…”

“…In nearly all of the published research on anorthoscopic perception to date 1 , 3 – 7 , the depicted figures have been 2-dimensional (2-D) objects, drawings, or contours, and they were simply translated back and forth behind the slit (with the slit and depicted figures being presented in a fronto-parallel plane). In 2021, Norman et al 2 found that observers can identify and discriminate the 3-dimensional (3-D) shapes of solid objects that rotate in depth behind narrow slits in an occluding surface, thus demonstrating that the human visual system can temporally integrate meaningless fragments of visible surfaces into coherent perceptions of 3-D object shape. The phenomenon discovered by Zöllner 3 thus generalizes from 2-D figures to the perception of solid (i.e., 3-D) object shape.…”

The importance of contours for visual object recognition and discrimination

“…For those object orientations that produced prominent convexities and concavities in the silhouette/shadow outer boundary contour, the participants' identification performance was excellent (see Figures 6 and 7 of Norman et al, 2000). In a very recent similar study, Norman, Dukes, Shapiro, Sanders, and Elder (2020) demonstrated that human observers do not necessarily need to see entire object silhouettes or cast shadows; if moving objects are presented, one can occlude or block over 90 percent of an object's silhouette/cast shadow and the object will remain highly recognizable.…”

Temporal integration characteristics of an image defined by binocular disparity cues

“…Some studies have investigated three-dimensional (3D) object recognition from the stimulus movement, although most slit-vision studies have focused on two-dimensional pictorial recognition of the images. Research has shown that shape perception is possible, even in the slit vision of a rotating object or a cube frame ( Norman et al, 2021 ). This indicates that even under slit-vision conditions, a participant is able to temporally integrate the local motion information to recognize and estimate the whole object's motion and restore the 3D shape.…”