The role of monocular regions in stereoscopic displays

Random-dot stereograms of an object standing out from a background always contain a monocular region at the side of the foreground object. This is equivalent to the monocularly occluded part of the background in the real-life viewing of one object in front of another. The role of these monocular regions in the stereoscopic process has not been investigated previously, although it is generally assumed that they are a source of difficulty in stereoscopic resolution because of the unmatchable texture within them. The basis of the present study was a prediction that the presence of texture within these regions would facilitate rather than retard stereoscopic processing. This prediction follows from a hypothesis that stereoscopic processing is initially located at disparity discontinuities. Unmatched regions are only found at such discontinuities, and could serve to locate them.     

Depth of monocular elements in a binocular scene: the conditions for da Vinci stereopsis

Quantitative depth based on binocular resolution of visibility constraints is demonstrated in a novel stereogram representing an object, visible to 1 eye only, and seen through an aperture or camouflaged against a background. The monocular region in the display is attached to the binocular region, so that the stereogram represents an object which is only partially visible to the eye that sees it. The results show that this feature is necessary for quantitative depth, which is not found for a fully visible monocular object in the same location, and that depth in these displays, although very precise, is not based on fusional Stereopsis. The findings provide clear support for the existence of a process of da Vinci Stereopsis, but one more sophisticated than the one proposed by K. Nakayama and S. Shimojo (1990).     

Depth from binocular half-occlusions in stereoscopic images of natural scenes

Over the past two decades psychophysical experiments have firmly established that binocular half-occlusions are useful sources of information for the human visual system. The existing literature has focused on simplified stimuli that have no additional cues to depth, apart from stereopsis. From this large body of work we can be confident that the visual system is able to exploit binocular half-occlusions to aid depth perception; however, we do not know if this signal has any influence on perception when observers view complex stereoscopic stimuli with multiple sources of depth information. This issue is addressed here with the use of stereoscopic images of natural scenes, some of which have been digitally altered to manipulate a major half-occlusion signal. Our results show that depth-ordering judgments for these relatively complex stimuli are significantly affected by the nature of the half-occlusion signal, but only when highly textured surfaces are viewed. Under such conditions, the replacement of a binocular half-occlusion with background texture slows reaction time relative to performance when the occluded region is consistent with the foreground object. This result is specific to conditions when the depth ordering is correct (ie not reversed) and depends upon the size of the half-occlusion. The influence of the half-occlusion information in the presence of potent depth cues such as perspective, texture gradient, shading, and interposition is convincing evidence that this information plays a significant role in human depth perception.     

Selective adaptation of monocular and binocular neurons in human vision

Psychophysical techniques were used to examine how subpopulations of visual neurons varying in their ocular dominance interacted in determining performance on a visual task. Using an asymmetric alternating adaptation of the left and right eyes, we manipulated the sensitivity of monocularly driven neurons while keeping the sensitivity of binocularly driven neurons constant. Relative threshold elevations were measured in the left eye, right eye, and both eyes of five observers following different ratios of alternating adaptation. It was found that whereas monocularly measured aftereffects varied monotonically as a function of the adaptation duration of the measured eye, the magnitude of the binocularly measured aftereffect remained constant regardless of how the adaptation was divided between the two eyes. This suggests that neurons differing in their ocular dominance pool their activity in determining sensitivity to a test target.     

Dark vergence in human infants: implications for the development of binocular vision

The resting position of binocular alignment in 1- to 4-month-olds, 6- to 18-month-olds, and adults was estimated by obtaining photographic measures of interpupillary distance in total darkness. The mean dark vergence position in the adult group corresponded to a distance of approximately 100 cm, similar to a previous estimate of 120 cm (Owens and Leibowitz 1976). The mean dark vergence position in the group of 1- to 4-month-olds was approximately 25 cm and in the group of 6- to 18-month-olds was approximately 50 cm. These near dark vergence positions in infants provide strong evidence that inaccuracies in binocular fixation during early infancy are not the result of a divergence bias. The developmental implications of dark vergence, accommodative vergence, and fusional vergence for the control of binocular alignment and distance perception during infancy are discussed.     

Binocular and monocular detection of Gabor patches in binocular two-dimensional noise

Contrast thresholds for detecting sine-wave Gabor patches in two-dimensional externally added random-pixel noise were measured. Thresholds were obtained for monocular and binocular signals in the presence of spatial correlated (identical) and uncorrelated (independent) noise in the two eyes. Measurements were obtained at four different spectral densities of noise (including zero). Thresholds were higher for monocular stimuli than for binocular, and higher in the presence of correlated noise compared to uncorrelated noise. The magnitude of binocular summation, similar in correlated and uncorrelated noise, decreased with increasing noise strength. The independent contributions of internal noise and sampling efficiency to detection were analysed. Sampling efficiencies were higher for binocular than for monocular viewing for both types of noise, with values being higher with uncorrelated noise. Binocular stimuli showed a lower equivalent noise level compared to the mean monocular case for both types of noise.     

Fragmentation of monocular afterimages in individuals with and without normal binocular vision

Monocular afterimages of a vertical line were generated in each eye of binocularly normal individuals and of those lacking binocular single vision or stereopsis as a consequence of childhood strabismus. Fragmentations and disappearances of the afterimages occurred less frequently in the strabismic than in the normal subjects; they were also less frequent in the dominant (nonamblyopic) eyes of the strabismics than in their amblyopic eyes. The afterimages remained visible for longer when presented to the dominant eyes of the strabismic subjects, and a similar effect was found between the sighting dominant and nondominant eyes of normal subjects.     

Temporal integration of monocular images separated in time: stereopsis, stereoacuity, and binocular luster

We evaluated stereopsis and binocular luster using electronically controlled shutter glasses with alternating monocular stimulation. In Experiment 1, we used the standard method for testing stereoacuity to obtain a gradual measure of stereopsis. Stereo thresholds decreased with increasing alternating frequency of two monocular half-images without a delay between them. Increasing delays led to increasing thresholds. In Experiment 2, we compared stereopsis resulting from two monocular half-images of a random-dot stereogram and binocular luster with respect to the minimum alternating frequency of the two half-images and the maximum interocular delay that were tolerated without a breakdown of the impression. Below 3 Hz, no stereopsis occurred. Binocular luster was observed only above 10 Hz. The mean threshold of interocular delay for detecting the global figure in a random-dot stereogram was about 51 msec, but for binocular luster it was about 20 msec. Overall, temporal integration was better for stereopsis than for binocular luster.     

Parallel working memory for spatial location and food-related object cues in foraging pigeons: binocular and lateralized monocular performance

During foraging, animals can increase their success by both remembering feeding sites and remembering food-related object cues. Because earlier studies have tested either the site or object memory in isolation, the aim of the present study was to evaluate how efficiently birds can utilize both memories simultaneously. Furthermore, the idea was tested that lateralization might be the principle of brain organization that allows for efficient parallel processing. Pigeons learned to search for food in a complex maze with 16 baited sites. To obtain the maximum reward they had to perform two tasks in parallel, a spatial working memory task and an object-specific working memory task. Birds performed well on this dual task but, compared with spatial working memory alone, they were impaired during the first choices of a trial (Experiment 1). When the left and the right brain hemispheres were tested separately by means of monocular occlusion (Experiment 2), object discrimination was better when birds used their right eye/left hemisphere. This was most pronounced during the first choices of a trial. On the spatial component of the task, performance on binocular trials was better than on monocular trials, but monocularly both hemispheres performed at the same level. Results show that on this dual task, discrimination of food-related object cues predominantly involved the left brain hemisphere whereas both hemispheres contributed equally to spatial performance.     

A comparison of monocular and binocular depth perception in 5- and 7-month-old infants

Monocular depth perception was compared with binocular depth perception in 5- and 7-month-old infants. Reaching was used as the dependent measure. Two objects, identical except in size, were presented simultaneously to each infant. The smaller object was within reach for the infants while the larger object was just beyond reach. The two objects subtended equal visual angles from the infants' observation point. With binocular presentation, 96% of the 7-month-olds' reaches and 89% of the 5-month-olds' reaches were for the nearer object. With monocular presentation, 58% of the 7-month-olds' reaches and 65% of the 5-month-olds' reaches were for the nearer object. The reaching preferences observed in the monocular condition indicated sensitivity to monocular depth information (motion parallax, accommodation, and relative size information were available). Binocular viewing, however, resulted in a far more consistent tendency to reach for the nearer object. This result suggests that the infants' perception of the objects' distances was more veridical in the binocular condition than in the monocular condition.     

Size-arrival effects: the potential roles of conflicts between monocular and binocular time-to-contact information, and of computer aliasing.

With computer simulations of self-motion, participants approached a floating object and tried to "jump" over it without collision. Participants "jumped" significantly later over small objects than they did over larger objects. This occurred when the displays were viewed monocularly or binocularly, a finding that suggests that such size-arrival effects (DeLucia & Warren, 1994) were not due to a conflict between monocular and binocular cues to time-to-contact (TTC) information (Tresilian, 1994, 1995). Moreover, the results further suggest that size-arrival effects are not due to irregularities in TTC information that can occur from computer aliasing and that the latter does not always affect TTC estimation; visual information used in such judgments does not seem to be extracted on a frame-by-frame basis.