Perception

Understanding the mechanisms underlying the multistability of reversible figures may provide valuable insights into the normal functioning of our visual system. The proposed factors that control the perceptual alternations of reversible figures can be classified into bottom–up and top–down processes. In the present study, we report differences in top–down effects on the reversal rate depending on whether a structural perspective (Necker cube, Schröder staircase) or a meaningful content (duck/rabbit figure, chef/dog figure) is subject to the reversal phenomenon. In order to activate top–down mechanisms explicitly the subjects had the instruction to bring the reversal rate under voluntary control. The results indicated that both slowing down and speeding up the rate of alternations was more effective for the content-reversal figures (duck/rabbit, chef/dog) than for the rather abstract perspective-reversal figures (Necker cube, Schröer staircase). In order to investigate the effect of meaningfulness in figure/ground reversals, the effect of the same instructional variable was also determined for Rubin's vase/faces and the Maltese cross. The results showed a similar tendency as in the case of the comparison between perspective reversals and content reversals. Possible cognitive processes that may play a role in top–down influences on figure reversal and theoretical implications of these findings for the interaction of bottom–up and top–down processes are discussed.

Outline-shape information may be particularly important in the recognition of depth-rotated objects because it provides a coarse shape description which gives first-pass information about the structure of an object. In four experiments, we compared recognition of silhouettes (showing only outline shape) with recognition of fully shaded images of objects, by means of a sequential-matching task. In experiments 1 and 2, the first stimulus was always a shaded image, and the second stimulus was either a shaded image or a silhouette. Recognition costs associated with a change in viewpoint were no greater for silhouettes than they were for shaded images. Experiments 3 and 4 replicated the design of the earlier experiments, but showed a silhouette as the initial stimulus, rather than a shaded image. In these cases, recognition costs associated with a change in viewpoint were greater for silhouettes than for shaded images. Combined, these results indicate that, while visual representations clearly include additional information, outline shape plays an important role in object recognition across depth rotation.

We report two experiments indicating that varying the configuration of face features changes perception of an oval aperture windowing the face: as the eyes and mouth of a frontal-view face photograph are moved vertically toward face boundaries, the oval appears increasingly elongated, taller, and narrower; when eyes and mouth are moved toward the nose, the oval appears increasingly rounder, shorter, and wider. This shape illusion is maximised when faces appear upright within the oval, and major face features (eyes, nose, and mouth) appear in their correct relative locations. These results establish that processing of a face configuration can affect perception of a geometric shape that shares visual space with a face. Whether the illusion is face-specific or a special case of a more general geometric illusion is discussed.

It is argued that the whole face is more dominant than the individual features. In the case of a jumbled face the external pattern is dominant when a face is upright, whereas the internal pattern is dominant when a face is inverted.

Under certain conditions, high-contrast moving figures induce adjacent illusory regions, ‘wakes’ and ‘spokes’, which have contrast polarity opposite the inducing figures. In this paper we document properties of these novel phenomena. When the illusions are induced by a moving bar, spokes appear on the side of the bar closer to fixation and connect the bar to the fixation point, regardless of the momentary position of the bar or whether it is moving to the left or to the right. Although spokes often extend up to the fixation point, they never extend beyond it. This is not due to blocking of the spoke's spread by the fixation point, because in another experiment spokes extend directly through an intervening figure. Whereas spokes emanate from the end of a horizontally moving bar closest to fixation, wakes emanate from the end farthest from fixation. In contrast to spokes, wakes do not show a towards-fixation bias. Instead, the wake's end trails the position of the bar, like a ship's wake. The higher the bar velocity, the more the end of the wake appears to trail it, suggesting that wakes are caused by a process which spreads from the edge of moving figures. Wakes and spokes, as distinct illusions, should provide significant constraints on theories of human motion and brightness perception processes.

Four experiments demonstrate that lines indicating path of movement can generate rotational percepts in a multistable motion display that usually produces only horizontal or vertical motion percepts. The properties of the path-of-movement lines are predicted by a neural-network theory of visual perception. Experimental results validate the theory's predictions by demonstrating that movement of the display elements seems to follow an increasing luminance gradient in lines but not bars, and that illusory contours have similar effects. Experimental results also demonstrate that, in a choice between movement along lines drawn parallel or orthogonal to possible motion paths, observers more often see movement along the lines parallel to the motion path. These results suggest modifications to current computational and neurophysiological theories of motion perception.

Motivated by the debate between indirect and direct theories of perception, a large number of researchers have attempted to determine whether judgments of time to collision are based on the ratio of perceived distance to perceived speed or on the ratio θ/(d θ/dt), ie tau. Despite the considerable research effort devoted to this question there seems to be no clear resolution. We used a staircase tracking procedure to estimate errors in estimating time to collision for a simulated approaching object. To investigate the role of perceived distance in the judgment of time to collision, we asked observers to alternate between two viewing distances (100 and 500 cm). For the 500 cm viewing distance, we magnified the visual display by a factor of five so that the retinal images [and the values of θ/(d θ/dt) through time] were identical for the two viewing distances. All visual cues to distance were available. There were no significant differences between estimates of time to collision made at the two viewing distances. We conclude that our observers ignored perceived distance when estimating time to collision and based their responses on θ/(dθ/dt). We concur with recent proposals that, in the future, time-to-collision research should move away from the either/or analysis of different information sources that has dominated previous studies towards investigations of how different information sources are integrated.

A study is reported of an exocentric pointing task in all three dimensions, in near space, with only two visible luminous objects—a pointer and a target. The task of the subject was to aim a pointer at a target. The results clearly show that visual space is not isotropic, since every set direction appeared to consist of two independent components—one in the projection onto a frontoparallel plane (tilt), the other in depth (slant). The tilt component shows a general trend across subjects, an oblique effect, and can be judged monocularly. The slant component is symmetrical in the mid-sagittal plane, requires the use of binocular information, and shows considerable differences between subjects. These differences seem to depend on the amount of binocular information used by each subject. There was a remarkably high level of consistency in the exocentric pointing, despite the absence of environmental cues. The within-subject consistency in the settings of the pointer corresponds to a consistency of about 1 min of arc in disparity of its tip, even though the pointer and target are separated by more than 5 deg.

We investigated the perceptual grouping of sequentially presented sounds—auditory stream segregation. It is well established that sounds heard as more similar in quality, or timbre, are more likely to be grouped into the same auditory stream. However, it is often unclear exactly what acoustic factors determine timbre. In this study, we presented various sequences of simple sounds, each comprising two frequency components (two-tone complexes), and measured their perceptual grouping. We varied only one parameter between trials, the intercomponent separation for some of the complexes, and examined the effects on stream segregation. Four hypotheses are presented that might predict the extent of streaming. Specifically, least streaming might be expected when the sounds were most similar in either (1) the frequency regions in which they have energy (maximum spectral overlap), (2) their auditory bandwidths, (3) their relative bandwidths, or (4) the rate at which the two components beat together (intermodulation rate). It was found that least streaming occurred when sounds were most similar in either their auditory or their relative bandwidths. Although these two hypotheses could not be distinguished, the results were clearly different from those predicted by hypotheses (1) and (4). The implications for models of stream segregation are discussed.