Perception

It has recently become apparent that if face images are degraded by spatial quantisation, or block averaging, there is a nonlinear acceleration of the decline in accuracy of recognition as block size increases. This suggests recognition requires a critical minimum range of object spatial frequencies. Two experiments were performed to clarify the phenomenon. In experiment 1, the speed and accuracy of recognition for six frontoparallel photographs of faces were measured. After familiarisation training sessions, the images were shown for 100 ms with 11, 21, and 42 pixels per face, horizontally measured. Transformations calculated to remove the same range of spatial frequencies were performed by means of quantisation, a Fourier low-pass filter, and Gaussian blurring. Although accuracy declined and speed increased in a significant, nonlinear manner in all cases as the image quality was reduced, it did so at a faster rate for the quantised images. In experiment 2, faces rated as being typical were shown at 9, 12, 23, and 45 pixels per face and with appropriate Fourier low-pass versions. The nonlinear decline was confirmed and it was shown that it could not be attributed to a ceiling effect. A further condition allowed quantised and Fourier low-pass conditions to be compared with an unstructured-noise condition of equal strength to that of the quantised images. These gave comparable, but slightly less impaired, recognition than the quantised images. It can be inferred from these results that the removal of a critical range of at least 8–16 cycles per face of information explains the step decline in recognition seen with quantised images. However, the decline found with quantised images is reinforced by internal masking from pixelisation.

One can shear a pattern of lines to produce an anomalous contour which has a perceptual influence similar to that of a straight line segment. Illusion effects have been found with configurations which contain these anomalous contours, as well as cross adaptation with respect to luminance contours. We have found that sheared-line contours will bias judgment of collinearity, ie perceived alignment, of a luminance contour. The angular induction effects are similar to those reported for interactions between luminance contours, and the same equation can be used to model both kinds of data. The results of this experiment support the neuroreductionist view that anomalous and luminance contours are processed at the same level of the nervous system. Additionally, we suggest that with both types of contour the perceptual system registers and responds to the alignment of local brightness differentials.

The redundant-signals effect describes the general phenomenon that simple reaction times to two simultaneously presented signals are typically faster than the corresponding reaction times to each of the signals presented alone. Recent studies (eg Miller 1982, 1986) indicate that models of probability summation in which an independent detection of both signals is assumed cannot completely account for the observed shortening of the reaction times. Therefore, models in which some kind of coactivation is assumed are often considered as an alternative explanation. In the present study simple reaction times to parallel lines are compared with those to orthogonal lines and single lines. Our first hypothesis is that because of the redundant-signals effect, the reaction time to configurations consisting of two lines (either parallel or orthogonal) will generally be faster than the reaction time to a single line. Furthermore, line detection can be related to orientation-specific line detectors. Therefore, parallel lines may be thought to activate similar line detectors and, by coactivation, facilitate detection. As our second hypothesis we thus expect that the reaction time to parallel lines will be shorter than the reaction time to orthogonal lines. To test these hypotheses, we conducted a simple reaction-time experiment in which signal onset asynchronies ranging from 0 to ±56 ms for the orthogonal lines were used. In addition, reaction times to parallel lines and single lines were measured. Both hypotheses are supported by our data. We formulate a stochastic model which is able to explain both statistical facilitation and coactivation in a physiologically plausible way.

The human visual system makes effective use of shading alone in recovering the shape of objects. Pictures of sculptures are readily interpreted—a situation where shading provides virtually the sole cue to shape. However, shading has been considered a poor cue to depth in comparison with retinal disparity and kinetic cues. Curvature discrimination thresholds were measured with the use of a surface-alignment task for a range of surface curvatures from 0.16 cm⁻¹ to 1.06 cm⁻¹. Weber fractions were around 0.1, demonstrating considerable precision in this task. Weber fractions did not vary substantially as a function of surface curvature. Rotation of the light source around the line of sight had no effect on curvature discrimination but rotation towards the viewer increased discrimination thresholds. In contrast, slant discrimination declined with rotation of the light-source vector towards the viewpoint. When a band-limited random grey-level texture was mapped onto the sphere, curvature discrimination thresholds increased gradually as a function of texture contrast, even though texture and shading provided consistent cues to depth. Adding texture also increased slant discrimination thresholds, demonstrating that texture can act as a source of noise in shape-from-shading tasks.

The psychophysical findings have been used to evaluate whether current algorithms for shape from shading in computer vision could serve as models of human three-dimensional shape analysis and to highlight low-level intramodular interactions between depth cues. It is demonstrated that, in the case of surfaces defined by shading, curvature descriptions are primary and do not depend upon the prior encoding of surface orientation, and Koenderink's local-shape index is suggested as an alternative intermediate representation of surface shape in the human visual system.

Pattern-acuity tasks have provided valuable information about the precision with which the visual system can make judgments about relative spatial position in two-dimensional images. However, outside the laboratory the visual system is habitually faced with the more difficult task of making positional judgments within a three-dimensional spatial environment. Thus our perceptual systems for representing surface shape also need to support the recovery of the location and disposition of features in a three-dimensional space. An investigation of the precision of three-dimensional position judgments in two spatial-judgment tasks, arc length bisection along geodesics and geodesic alignment, is reported. The spatial-judgment tasks were defined with reference to a sphere rendered by means of ray-casting techniques. The presence of shading and texture cues had no effect on discrimination thresholds in either task. Observers' constant errors were generally less than the just noticeable distance, demonstrating that the observers can perform these positional judgment tasks without substantial bias. It is argued that there is no explicit computation of arc length on the basis of shading and texture information and that surface-orientation information cannot be used as a reference in geodesic-alignment tasks. The results raise questions about the utility of a representation of surface orientation in the human visual system.

First it was established that the size illusion occurs not only when complex patterns are presented kinaesthetically but even with simple stimuli such as straight lines and circles. Thus it was established that information overload is not the underlying cause of the illusion. The size illusion was investigated in children aged from five to twelve years and in adults. The stimulus circle was presented in passive kinaesthetic, sequential visual, in combined kinaesthetic and sequential visual, in static visual, and in combined kinaesthetic and static visual conditions either at fast or at slow speeds. A between-subjects design was used. In all conditions where the stimulus presentation included the kinaesthetic modality alone or in combination with visual or sequential visual conditions, size overestimation was found, while in the sequential visual and static visual conditions overestimation of the size of the stimulus was absent. Further, the kinaesthetic illusion was stable across the age range tested, indicating that the illusion is a property of the kinaesthetic system and is not influenced by learning, ie is hard wired.

Binocular eye movements were measured in subjects experiencing the wallpaper illusion. It was found that a physical displacement of the fixation point by more than 1 m out of the plane of apparent localisation of the strips had no influence on the illusory location of these strips. Hence, illusory localisation in the wallpaper illusion is independent of the actual magnitude of the subject's convergence angle, once the illusion has come into existence. This result suggests that convergence does not serve as a source of information about apparent distance, at least in the wallpaper illusion.

It has long been accepted that amongst patterns which are bilaterally symmetrical, those which have their axis of symmetry vertical are more saliently symmetrical than patterns whose axis of symmetry is at some other orientation. The evidence regarding the relative salience of other orientations of axis of symmetry is somewhat more equivocal. In experiment 1, subjects were required to discriminate between symmetric or random-dot patterns when the axis of symmetry was at one of eighteen different orientations, spaced 10° apart, both clockwise and counterclockwise of vertical to horizontal. The data indicated that vertical was most salient, then horizontal but that, unlike in the classical oblique effect for contrast sensitivity, performance for precisely diagonal axes was better than that for surrounding axis orientations. Additional data (from experiments 2 and 3) also showed that the salience of vertical and horizontal axes of symmetry can be manipulated extensively by varying the range of stimuli presented, presumably by manipulating the scanning or attentional strategy adopted by the observer. Many previous studies of symmetry perception may have confounded hard-wired salience for vertical symmetry with scanning or attentional strategies.

The observer looked for a target pattern differing from distractors in orientation at one spatial scale only (either at a global or at a local scale) and ignored the other. The stimulus patterns in the search array were vertical or horizontal bars consisting of four oblique line segments sharing the same orientation (45° or 135°). In the search at the global scale, the target and distractors differed from each other in the bar orientation, but not in the orientation of the line segments, which was random. In the search at the local scale, the observer had to use the line orientation for discriminating the target and distractors (the bar orientation was random). The results showed that even though the search was parallel at both scales (ie the search time did not increase with an increasing number of distractors), target detection at the local scale required considerably more time than at the global scale. This latter finding is in agreement with the phenomenon of ‘global precedence’.