Quantitative perceived depth from sequential monocular decamouflage.

We present a novel binocular stimulus without conventional disparity cues whose presence and depth are revealed by sequential monocular stimulation (delay > or = 80 ms). Vertical white lines were occluded as they passed behind an otherwise camouflaged black rectangular target. The location (and instant) of the occlusion event, decamouflaging the target's edges, differed in the two eyes. Probe settings to match the depth of the black rectangular target showed a monotonic increase with simulated depth. Control tests discounted the possibility of subjects integrating retinal disparities over an extended temporal window or using temporal disparity. Sequential monocular decamouflage was found to be as precise and accurate as conventional simultaneous stereopsis with equivalent depths and exposure durations.

Perspective based on stereopsis and occlusion.

Perspective is usually considered a monocular pictorial cue, distinct from other cues such as occlusion and stereopsis. We cut across these distinctions by asking whether purely binocular (cyclopean) contours, created by stereoscopically shifting a region of homogeneous texture nearer or further than its surround, can act as a linear-perspective cue and whether the contours' ability to do this is influenced by their surface belongingness. We found that the left/right orientation of cyclopean trapezoids nearer than a surround strongly influenced perceived slant, showing that perspective constraints are applied to stereoscopically derived contours. Further regions, however; appeared as surfaces seen through a trapezoidal aperture. Because the aperture "owned" the trapezoidal contours, their orientation had little effect on perceived slant. We conclude that the application of perspective constraints depends critically on how contours are classified by stereo-specified occlusion relationships among surfaces and that perspective, stereopsis, and occlusion are not distinct processing systems.

Global-perspective jitter improves vection in central vision.

Previous vection research has tended to minimise visual-vestibular conflict by using optic-flow patterns which simulate self-motions of constant velocity. Here, experiments are reported on the effect of adding 'global-perspective jitter' to these displays--simulating forward motion of the observer on a platform oscillating in horizontal and/or vertical dimensions. Unlike non-jittering displays, jittering displays produced a situation of sustained visual-vestibular conflict. Contrary to the prevailing notion that visual-vestibular conflict impairs vection, jittering optic flow was found to produce shorter vection onsets and longer vection durations than non-jittering optic flow for all of jitter magnitudes and temporal frequencies examined. On the basis of these findings, it would appear that purely radial patterns of optic flow are not the optimal inducing stimuli for vection. Rather, flow patterns which contain both regular and random-oscillating components appear to produce the most compelling subjective experiences of self-motion.

Failure to detect changes in color for lines rotating in depth: the effects of grouping and type of color change.

A new technique for measuring change detection was introduced in which contours rotating in depth around a vertical axis (in a computer display) could be altered in color as they passed through their point of minimum extension (the median plane) where a thin static vertical occluder hid the change. Sets of five or six contours were either strongly grouped (similar in length, orientation and spacing) or weakly grouped (of variable length, orientation and spacing). Changes consisted of one line changing to a new color or else two lines swapping colors. The measure was the proportion of missed changes. When subjects were not instructed to look for change almost no changes were reported although subjects were told beforehand that they would have to describe the configuration after viewing it. When subjects were instructed to look for changes, it was found that detection of color change was significantly better for strongly grouped lines. It is proposed that grouping, by reducing redundancy, also reduces attentional demands with respect to the properties on which it is based, making it easier to attend to and therefore detect changes in other properties. We found that it was much easier to detect the introduction of a new color than to detect a swap between two existing colors. It is hypothesized that swap-type changes were harder to detect because they required attention to a conjunction of position and color.

Stereopsis based on monocular gaps: metrical encoding of depth and slant without matching contours.

It is often the case in binocular vision that one eye can see between two objects lying at different distances but the other eye cannot. We have found that the visual system is able to correctly interpret images produced this way in which a single solid rectangle in one eye is fused with two half-sized rectangles in the other eye separated by a vertical gap comprising the background. Two rectangles in depth are seen. It is as if the solid rectangle is treated as two components which each match one of the physically separated rectangles in the contralateral eye. The sign of the depth depends on which eye's view has the gap and its magnitude increases with gap width. Measured depth is found to be equivalent to real stereoscopic depth with a relative disparity equal to the monocular gap. If overall disparity differences are eliminated, between the left and the right images, variations in perceived slant of the two rectangles are still seen with increasing gap size. That two surfaces can be seen in metric binocular depth despite complete camouflage of their separation in one eye's view, suggests that stereopsis be regarded as a broad process of surface recovery not necessarily requiring image disparity at the location of the depth step.

Quantitative depth for a phantom surface can be based on cyclopean occlusion cues alone.

Liu, L., Stevenson, S.B., and Schor, C.M. (1994, Nature, 367, 66-669) reported quantitative stereoscopic depth in a phantom rectangle which appeared to lack conventional matching elements. Later, Gillam, B.J. (1995, Nature, 373, 202-203) and Liu, L., Stevenson, S.B., and Schor, C.M. (1995, Nature, 373, 203) and Liu, L., Stevenson, S.B., and Schor, C.M. (1997, Vision Research, 37(5), 633-644) indicated that the varying depth of the phantom rectangle could be based on stereoscopic matching. To remove the contaminating effects of conventional stereopsis from the Liu et al. (1994) original example, we presented a pair of parallel vertical lines to each eye where there is a central gap in the right line for the left eye's view and in the left line for the right eye's view. Observers saw a phantom rectangle bounded by subjective contours whose depth increased with the thickness of the lines. We attribute the quantitative variation of depth to a purely cyclopean (binocular) process sensitive to the pattern of contour presence and absence in the two eye's view.

Surface separation decreases stereoscopic slant but a monocular aperture increases it.

When an isolated surface is stereoscopically slanted around its vertical axis, perceived slant is attenuated relative to prediction, whereas when a frontal-plane surface is placed above or below the slanted surface, slant is close to the predicted magnitude. Gillam et al (1988 Journal of Experimental Psychology: Human Perception and Performance 14 163-175) have argued that this slant enhancement is due to the introduction of a gradient of relative disparities across the abutment of the two surfaces which is a more effective stimulus for slant than is the gradient of absolute disparities present when the slanted surface is presented alone. To test this claim we varied the separation between the two surfaces, along either the vertical or depth axis. Since these manipulations have been reported to reduce the depth response to individual relative disparities, they should similarly affect any slant response based on a gradient of relative disparities. As predicted, increasing the separation, vertically or in depth, systematically reduced both the perceived slant of the stereoscopically slanted surface and also the stereo contrast slant induced in the frontal-plane surface. These results are not predicted by alternative accounts of slant enhancement (disparity-gradient contrast, normalisation, frame of reference). We also demonstrated that sidebands of monocular texture, when added to equate the half-image widths of the slanted surface, increased the perceived slant of this surface (particularly when presented alone) and reduced the contrast slant. Monocular texture, by signalling occlusion, appeared to provide absolute slant information which determined how the total relative slant perceived between the surfaces was allocated to each.

Early demonstrations of subjective contours, amodal completion, and depth from half-occlusions: "stereoscopic experiments with silhouettes" by Adolf von Szily (1921).

If all-black figures are used, certain monocular appendages to binocular shapes are seen in depth, either nearer (when in a medial position) or further (when in a lateral position) than the binocular shape itself. These appendages also link to form subjective contours in front of the binocular shape or amodal completions behind it. These and other discoveries by von Szily, made before 1921, anticipate a number of modern findings.

Stimulus eccentricity and spatial frequency interact to determine circular vection.

While early research suggested that peripheral vision dominates the perception of self-motion, subsequent studies found little or no effect of stimulus eccentricity. In contradiction to these broad notions of 'peripheral dominance' and 'eccentricity independence', the present experiments showed that the spatial frequency of optic flow interacts with its eccentricity to determine circular vection magnitude--central stimulation producing the most compelling vection for high-spatial-frequency stimuli and peripheral stimulation producing the most compelling vection for lower-spatial-frequency stimuli. This interaction appeared to be due, in part at least, to the effect that the higher-spatial-frequency moving pattern had on subjects' ability to organise optic flow into related motion about a single axis. For example, far-peripheral exposure to this high-spatial-frequency pattern caused many subjects to organise the optic flow into independent local regions of motion (a situation which clearly favoured the perception of object motion not self-motion). It is concluded that both high-spatial-frequency and low-spatial-frequency mechanisms are involved in the visual perception of self-motion--with their activities depending on the nature and eccentricity of the motion stimulation.

Gradients as visual primitives.

Following J. J. Gibson (1950), it is implicitly assumed in the literature that texture gradients are directly available as perceptual primitives. Yet, the depth response to compression gradients is poor compared with gradients of linear perspective. This may indicate that mechanisms for directly detecting the differential structure that constitutes a compression gradient do not exist. We tested this hypothesis outside the context of depth perception by measuring the speed with which participants could detect a gradient anomaly as a function of the number of elements in the gradient. Only in the case of linear perspective did anomalies "pop out." This was attributable to the emergent feature of alignment of the ends of the elements forming the gradient rather than the direct detection of its differential structure. It is argued that gradients are not perceptual primitives and that the poor depth response to compression in a variety of context (motion parallax, stereo, and perspective) therefore is not surprising.

Upregulation of actin gene expression in cells expressing exogenous beta-amyloid precursor protein.

We investigated the effects of beta-amyloid peptide precursor (APP) overexpression upon the levels of other mRNAs. Using quantitative slot-blot hybridization and immunoblot analysis we observed that enhanced levels of APP elevated the levels of beta-actin and beta-actin mRNA. Our results also suggest that the cytoplasmic domain of APP is crucial for the elevation in beta-actin gene expression.

Suppression of Alzheimer amyloid precursor protein (APP) expression by exogenous APP mRNA.

To establish a cell line expressing enhanced levels of beta-amyloid precursor protein (beta-APP), we constructed plasmid DNAs expressing beta-APP-751 mRNA and transfected them into COS-1 cells. Using a modified version of the reverse transcriptase polymerase chain reaction which is RNA sensitive to study the beta-APP iso-RNAs, we have made the unexpected observation that enhanced expression of beta-APP-751 mRNA resulted in a significant reduction of beta-APP-770 and -695 mRNA levels. Suppression of beta-APP-770 and -695 was also observed in cells expressing truncated and chimeric beta-App-751 mRNAs. Similar observations were made in P19 cells expressing a chimeric beta-APP-751 mRNA where endogenous beta-APP-751 mRNA levels also were decreased. Also, suppression of beta-APP-770 and -751 mRNAs was observed in human kidney cells expressing exogenous beta-APP-695 mRNA.

Panum's limiting case: double fusion, convergence error, or 'da Vinci stereopsis'.

Panum's limiting case--a perceived depth difference between two lines in one eye combined with only one in the other eye--has long been considered weak or reversible. In the last few years this has led to strong promotion of the view that Panum's case is not based on a stereoscopic process involving double fusion, that only one line is fused, with the depth of the other one attributable either to fixation disparity or to occlusion cues. This view is refuted in two ways. First it is shown that when the separation of the two lines, considered as a disparity, is within the range of 'patent stereopsis', depth is perceived with a precision and accuracy indistinguishable from regular stereopsis. The predictions of nonstereoscopic theories concerning the effects of fixation are not borne out at small disparities. Second, very compelling Panum versions of orientation and curvature disparity are introduced, which are difficult to account for except by a process of double fusion. Finally it is shown that at large disparities the depth in Panum's case deviates from prediction more frequently than does regular stereoscopic depth.

Cue conflict and stereoscopic surface slant about horizontal and vertical axes.

The way in which a planar surface is defined or configured may affect its apparent slant about a given axis, and the magnitude of slant-axis anisotropies. The authors have previously suggested that (i) these within-axis and between-axis configuration effects may be attributable, in part at least, to the perspective-disparity conflict generated when geometrically frontoparallel configured surfaces are slanted stereoscopically; and (ii) that implicit contours, defined by line endings or conjunctions, may have effects analogous to those seen with explicit contours. These possibilities were directly examined in two experiments. In experiment 1, slant-axis anisotropy was progressively induced by adding horizontal lines to a vertical-line (zero anisotropy) grid under conditions of cue conflict; slants about the vertical (but not the horizontal) were attenuated-demonstrating a clear and systematic nexus between surface configuration and slant-axis anisotropy. The presence of regular implicit horizontals similarly and selectively attenuated the slant perceived about the vertical. In experiment 2, cue conflict was seen to exacerbate slant-axis anisotropy, but clearly could not fully account for it. There was an axis asymmetry in the effect of degrading implicit contours: degradation had a marked impact on perceived slant about the horizontal but not the vertical axis.

A proximity-contingent stereoscopic depth aftereffect: evidence for adaptation to disparity gradients.

Prolonged inspection of a surface slanted in the third dimension of visual space typically results in a negative aftereffect such that, after adaptation, a surface in the fronto-parallel plane will appear slanted in the opposite direction. Binocular disparity is not necessary to generate such effects, since they can be obtained monocularly, presumably via adaptation to texture gradient. Six experiments demonstrated durable stereoscopic depth aftereffects in the absence of a texture gradient--by using discrete disparate objects rather than slanted surfaces--and demonstrated that adaptation was to the interobject disparity gradient rather than to the relative disparity of the objects per se. The disparity required to null the obtained aftereffects was inversely proportional to the horizontal separation of elements, for a constant disparity, and directly proportional to the separation of subsequently presented probes. When elements differed in depth (disparity), but were not laterally separated, nulling disparity was significant but invariant with changes in the horizontal separation of probe elements. In that case, adaptation was (i) either to the disparity gradient generated by the vertical separation of probe elements (of which the relative disparity component was tapped); or (ii) to relative disparity per se.

Stereoscopic slant reversals: a new kind of 'induced' effect.

Data are presented from three experiments confirming an earlier finding that the stereoscopic slant perceived may be opposite to the geometrically predicted direction of slant (Gillam 1967). The stimulus for stereoscopic slant was created by imposing a disparity gradient on a frontal plane surface. Reversals are shown to occur readily for slants around a vertical axis but rarely for slant around a horizontal axis. Reversal frequency is greater for surfaces which have a regular pattern, providing good perspective information about slant. Cue conflict cannot explain reversals because adherence to perspective information predicts a perception of zero slant rather than reverse slant. A new explanation has been proposed attributing reversals to the ambiguity of horizontal disparity gradients and disambiguation of the disparity gradient by its relationship to the perspective gradient. It is shown that for any given disparity gradient there is a physical surface which would give rise to a slant reversed with respect to that normally predicted. Such a surface is eccentric in the field of view, with eccentricity given by the difference between the slants signalled by the disparity gradient and the perspective gradient. This explains why reversal responses to disparity gradients occur in the presence of perspective. It is proposed, on the basis of this analysis and the fact that reversals occur, that, like convergence and vertical disparity, perspective is a factor contributing to the correct scaling of disparity gradients in the horizontal meridian with respect to surface eccentricity.

Perspective, orientation disparity, and anisotropy in stereoscopic slant perception.

Stereoscopic depth estimates are not predictable from the geometry of point disparities. The configural properties of surfaces (surface contours) may play an important role in determining, for example, slant responses to a disparity gradient, and the marked anisotropy in favour of slant around a horizontal axis. It has been argued that variation in slant magnitude are attributable to the degree of perspective conflict present and that anisotropy is attributable to orientation disparity, which varies with the axis of slant. Three experiments were conducted in which configural properties were varied to try and tease apart the respective roles of orientation disparity and conflicting perspective in determining stereoscopic slant perception and slant axis anisotropy. The results could not be accounted for by the magnitude of the orientation disparities present. Conflicting perspective cues appeared to play a role but only for slant around a vertical axis. It was concluded that there are important configural effects in stereopsis attributable neither to orientation disparity nor to perspective.

Motion capture by a frame: global or local processing?

The possibility that frames serve to capture lines within them so that they appear on a coplanar surface was investigated, using coherence in direction of rotary motion (Gillam, 1972) as a quantitative measure of the coplanarity of frame and internal lines. It was found that perceptual coherence between a pair of lines was greatly increased by surrounding them with a frame, if the frame was perspectivally consistent with the lines. A second experiment showed that this grouping can be attributed to a global effect of the frame and cannot be accounted for by local grouping of the internal lines with components of the frame.

Orientation disparity, deformation, and stereoscopic slant perception.

Koenderink and van Doorn's theory, that the basis of stereoscopic slant perception is the deformation component of the disparity, field, was tested for slant around a horizontal axis, which produces images with a vertical ramp of horizontal disparity (horizontal shear) characterised by a global orientation disparity at the vertical meridian. The disparity field in this case can be parsed into two components, deformation and curl, which each contribute half of the orientation disparity. This case was compared with similar random-dot stimuli in which the deformation component was doubled and the curl component eliminated or vice versa. All three types of stimuli had identical orientation disparity at the vertical meridian. A condition in which there was no such orientation disparity, but deformation was present, was also included. It was found that perceived slant was not related to the deformation present, as Koenderink and van Doorn's theory would predict, but was predictable from the orientation disparity at the vertical meridian per se.