The Effect of Surface Orientation on the Perception of Stereoscopic Corrugations

For most observers there is a pronounced orientational anisotropy in the perception of three-dimensional corrugated surfaces (Rogers and Graham, 1983 Science221 1409 – 1411; Bradshaw and Rogers, 1993 Perception22 Supplement, 117). Low-frequency corrugations which are oriented vertically have been found to have higher disparity modulation thresholds, the amount of perceived depth at suprathreshold levels is smaller, and typically they take longer to see than horizontally oriented corrugations. In the present experiments, the orientation of the corrugations was manipulated (from horizontal to vertical in 22.5 deg increments) to investigate the effect of surface orientation on both (i) a threshold detection task and (ii) a suprathreshold depth-matching task. The stimuli were 10 deg in diameter and were presented on two 12 inch monochrome monitors arranged to form a Wheatstone stereoscope. The surfaces were modulated in depth at four different corrugation frequencies (from 0.1 to 0.8 cycle deg−1 in octave steps). Thresholds were found to increase monotonically with increasing surface orientation from the horizontal: ∼2.5 arc s for horizontal corrugations to ∼10 arc s for vertical corrugations. The increase in thresholds was less marked for surfaces with higher corrugation frequencies. The rate of increase of threshold was greatest for surface orientations beyond 45°. A different pattern of results was found in the suprathreshold depth-matching task. Although the perceived depth in vertically oriented corrugations was significantly smaller (>50%) than for horizontally oriented corrugations, the largest amount of perceived depth was found for corrugated surfaces oriented at 45°. These results suggest that the disparity information used to process stereoscopic corrugations at threshold may be different from that used to process suprathreshold surfaces.

Further evidence that Whorfian effects are stronger in the right visual field than the left

The Whorf hypothesis holds that differences between languages induce differences in perception and/or cognition in their speakers. Much of the experimental work pursuing this idea has focused on the domain of color and has centered on the issue of whether linguistically coded color categories influence color discrimination. A new perspective has been cast on the debate by recent results that suggest that language influences color discrimination strongly in the right visual field but not in the left visual field (LVF). This asymmetry is likely related to the contralateral projection of visual fields to cerebral hemispheres and the specialization of the left hemisphere for language. The current study presents three independent experiments that replicate and extend these earlier results by using different tasks and testing across different color category boundaries. Our results differ in one respect: although we find that Whorfian effects on color are stronger for stimuli in the right visual field than in the LVF, we find that there are significant category effects in the LVF as well. The origin of the significant category effect in the LVF is considered, and two factors that might account for the pattern of results are proposed.

Perception of Slope in Photographs

Perceptions of characteristics of space such as slope, distance, and depth are frequently inaccurate, both in the real world and in pictures. We carried out experiments to study factors that influence the accuracy of perceived slope in photographs. Slopes varied in angle from 5° to 45° inclinations against the horizontal, and in the information available to the observer (outline shape and texture characteristics). We found that perceived slope is correlated with real slope ( r=0.99), but that observers consistently overestimate slope. The latter depends not only on the available information, but also on the focal length of the lens with which slopes were photographed. Overestimation is less pronounced for the wide-angle lens compared to the standard lens. A comparison of free viewing and viewing from the correct station-point showed that the latter leads to less overestimation of slope. Since the viewing distance was too far under free viewing, the results are compatible with geometrical optics. In a further experiment the effects of magnification and minification were studied by deliberately viewing the photographs from fixed points closer or further away than the station-point; this led to an increase and decrease in overestimation, respectively. Finally, results are frequently dependent on task characteristics: magnitude judgements of photographs without an anchoring point can only be accurate to a level of scale. Thus using an action-based matching task may lead to more accurate slope perception. We therefore carried out a comparison experiment using a matching task to check for the generality and action-dependence of our results. Practical implications for the use of photographs as surrogates for natural viewing are discussed.

Do Linguistic Categories Affect Colour Perception? A Comparison of English and Turkish Perception of Blue

Cultural relativists adduce the variation in colour categories across languages as prima facie evidence for linguistic relativity (language affects thought). However, there have been very few experiments that have gone beyond this observational level to assess the extent and the nature of linguistic differences on colour categorisation and perception. Here, we report experiments comparing English and Turkish speakers using a colour-grouping task and same - different tasks aimed at redressing this lack. Turkish categorises the blue region with two basic colour terms (lacivert ‘dark blue’ and mavi ‘blue’) whereas English has a single basic term. In experiment 1 subjects sorted a representative set of 65 colours into groups on the basis of their perceptual similarity. Native Turkish speakers were significantly more likely than native English speakers to form two distinct blue groups corresponding to the two basic blue terms of Turkish. In the same - different tasks we sought for possible categorical effects: enhanced discrimination across category boundaries and/or reduced discrimination within categories. For successive presentation, Turkish speakers were more accurate than English speakers in judgments of colour pairs that fell on opposite sides of the lacivert - mavi boundary. However, for simultaneous presentation, there was no difference between the two language groups. The results suggest that there are detectable effects of linguistic categories on colour cognition, but the locus of the effect may be in memory rather than perception.

What Spatial Frames of Reference are Used to Guide Off-Axis Aiming?

What spatial frames of reference are used to guide body centred actions such as pointing and aiming? Do observers learn to use body-scaled frames of reference with specific points of origin (eg hip vs shoulder)? Do they extrapolate general information about body scaling? Do they learn about the position of the body (and its individual parts) in absolute space? We investigated these questions using two kinds of aiming: off-axis aiming vs along-axis aiming. Subjects aimed a ‘gun’ (a laser pointer mounted at the end of a stick with a button at the other end) at various targets under three conditions. They practised aiming in one position for 60 ‘shots’ and then changed to a new position and completed another 60 shots. Transfer across position of ‘gun’ in absolute space, across position of body in absolute space, and across position of gun relative to body were assessed. Results suggested that off-axis aiming was controlled by a closed-loop error correction system requiring sight of results for learning to occur. Further, changes in the position of the body in absolute space (from kneeling to standing or vice versa) led to the worst aiming performance. Changes in gun position relative to the body or in absolute space had less effect on aiming accuracy, which suggests that it is not dependent on specific learned sensory - motor linkages. On the basis of these results we propose a preliminary model for off-axis training.

Far Distance Perception

Determining how people perceive distance is a central issue in the study of visual perception. Whilst near-distance perception has been extensively researched, far-distance perception has received little attention. We review the literature and illustrate key points with new data, focusing on analysis of available information/cues, degree of accuracy, measurement procedures, perceptual/cognitive influences, and geometry of perceived space. Analysis of available cues suggests that under natural viewing there is sufficient information for veridical distance perception. However, distance judgements are inaccurate and vary with mode of measurement (eg absolute vs relative judgements). Inaccuracy includes compression of distance. This has been modelled with the use of different geometries—most commonly power functions. Exponents often average 1.0 but this conceals considerable individual variation (in our data individual exponents ranged from 0.5 to over 1.0). Further, even for averaged exponents values vary between 0.8 and 1.25, as a function of viewing conditions, experimental method, and of the relative contribution of cognitive and perceptual factors. Evidence suggests that distance is encoded at an ordinal level, but for many practical tasks the final judgement must be metric (eg range finding) and this transformation is error-prone (in our data numerical estimates could be a factor of 10 out). Further, many natural judgements require perception of the full layout of the scene (including exocentric distances) rather than the more commonly investigated perception of egocentric distance. Evidence suggests that training based on practice with feedback produces some improvement in accuracy, but this is highly context-specific.

On the Validity of Colour Vision Tests in the Field: A Practical Assessment

We found a surprisingly high incidence of tritan-like (blue - yellow) symptoms in population samples from southern Africa. However, these results could be false positives because the test (The City University Colour Vision Test: CUCVT) was not administered under the prescribed illuminant (CIE Illuminant C). We therefore assessed the robustness of the test in order to estimate the true tritan rates in our data and to assess its usefulness under field conditions. First, we administered the test to normal young people under three illuminances (600, 300, and 150 lux), and three spectral distributions (C, reduced short wavelength, and reduced long wavelength). In addition, two viewing distances were used: the standard (14 inches) and double the standard. At the normal viewing distance, no errors were induced by reducing the illuminance or by changing the spectral distribution. However, at 28 inches, both illuminance and spectral changes induced tritan errors. Second, we assessed a sample of old people (over 65), first, under the prevailing illuminant; second, under Illuminant C; and third, under Illuminant C with increased intensity. In all cases, the old people in the sample made large numbers of tritan responses, but the frequency of tritan errors decreased under Illuminant C and decreased even more under enhanced illuminance. Thus, for normal young people at the prescribed distance the test is robust, but at increased distance the test is vulnerable to deviations from the prescribed illuminant. Further, older people can present as having tritan-like problems unless the illuminance is increased above the standard.