A fresh look at interocular grouping during binocular rivalry

During binocular rivalry, observers sometimes perceive one complete visual object even though component features of that perceptually dominant object are distributed between the two eyes and are in rivalry against other, dissimilar features. This interocular grouping cannot be explained by models of rivalry in which one eye or the other is completely dominant at any given moment. But perhaps global interocular grouping is achieved by simultaneous local eye dominance, wherein portions of one eye's view and complementary portions of the other eye's view become dominant simultaneously. To test this possibility, we performed two experiments using relatively large, complex figures as rival targets. In one experiment we used an "eye-swap" technique to confirm that within given, local spatial regions of rivalry it was the region of an eye--not a given stimulus feature--that was usually dominant. In a second experiment, we measured dominance durations for multiple, local zones of rivalry and then created 1-min animations of a global "montage" in which dominance within local regions was governed by the distributions of dominance measured empirically. These animations included significant periods of time during which global interocular grouping was evident; observers viewed these animations intermixed with actual rivalry displays, and the resulting tracking data confirmed the similarity in global dominance of the two display types. Thus interocular grouping during rivalry does not rule out local, eye-based rivalry, although synergistic and top-down influences almost certainly provide additional force in the promotion of interocular grouping.     

Stimulus specificity in spatially-extended interocular suppression

In typical binocular rivalry demonstrations, disparate images presented in corresponding locations to the two eyes are found to alternate perceptually over time. Alternation in perception can occur even if the images presented to the two eyes do not overlap, if they are sufficiently close in space. This implies a spatial spread in the interocular interaction. The current set of experiments explores how the luminance pattern of a target, in relation to a rivalrous suppressor, affects its susceptibility to suppression. It was found that the susceptibility to suppression of a target pattern was nonlinearly related to the amount of luminance variation along the target in the direction perpendicular to the suppressing stimulus. For instance, there was a strong effect of the orientation of the grating pattern within the target on the total time of suppression, with much more suppression for horizontal gratings than vertical gratings when suppressor bars were oriented vertically, regardless of the luminance pattern within the suppressors. Furthermore, it was shown that the inclusion of a spatial gap between the vertical suppressors and the central portion of the target does more than simply change the spatial relationships, it adds new figural information, such as vertically orientated edges in the targets, that modify the susceptibility to suppression of the target, thereby interfering with measurements of spatial interaction functions. All of the results are consistent with selectively suppressing stimulus information that would interfere with stereoscopic matching to aid the binocular fusion of disparate retinal images.     

Local binocular fusion is involved in global binocular rivalry

We examined whether interocular inhibition in binocular rivalry could occur at the interocular intersection of horizontal and vertical rectangular patches which are locally fusible but globally rivalrous between the two eyes. We measured contrast increment (and decrement) thresholds of a monocularly presented probe which was presented on the horizontal patch corresponding to the intersection. We found that the threshold was higher when the horizontal patch was perceptually suppressed than when it was dominant. In addition, threshold elevation did not occur when both patches were dominant, or when the horizontal patch was viewed in isolation. These results indicate that interocular inhibition occurs at the potentially fusible region, and the determination of binocular fusion or binocular rivalry does not depend on physical stimulus but rather perceptual state at the time.     

Temporal frequency and contrast tagging bias the type of competition in interocular switch rivalry

The nature of competition underlying perceptual alternations in binocular rivalry remains controversial. Interocular swapping of rivalrous stimuli can result in either slow irregular perceptual alternations that bridge multiple interocular switches or fast regular alternations that are time locked to the stimulus exchanges. We labeled either the inputs to the eyes or the individual rivalrous stimuli using temporal frequency and contrast tagging. Tagging of eye-of-origin signals enhanced the fast regular perceptual alternations associated with eye rivalry, while stimulus tagging shifted perception towards slow irregular alternations characteristic of stimulus rivalry. Thus, the type of competition in binocular rivalry can be biased based on additional cues in the visual inputs. The results are consistent with a model in which the brain combines information across multiple visual features to resolve ambiguities in visual inputs.     

Visual grouping on binocular rivalry in a split-brain observer

We studied the effects of visual grouping on binocular rivalry in the left and right hemispheres of a split-brain observer, JW. In Experiments 1 and 2, we compared responses to traditional rivalry stimuli (e.g., a red vertical grating presented to the left eye and a green horizontal grating presented to the right eye) with responses to Diaz-Caneja stimuli (i.e., half of each grating was presented to one eye and the other half to the other eye). As found for intact-brain observers, JW reported episodes of exclusive visibility of coherent stimuli (e.g., of a red vertical grating alternating with a green horizontal grating) with Diaz-Caneja stimuli that were fewer and briefer than with traditional stimuli. This occurred in both hemispheres, demonstrating that during binocular rivalry, contours from one eye can be grouped with those of the opposite eye to create a coherent percept, even in the isolated hemispheres of the split-brain observer. In Experiment 3, we studied the tendency of rivalry in adjacent patches to synchronize. When both patches were in one of JW's hemifields, rivalry synchronized for similarly oriented stimuli, the same as happened for intact-brain observers. When the patches were in JW's opposite hemifields, there was no synchronizing of rivalry, unlike what happened for intact-brain observers. This suggests that rivalry processed in JW's two hemispheres is independent. We conclude that rivalry is processed fully within each hemisphere.     

Feature-based activation and suppression during binocular rivalry

In the past decade, effects of pattern coherence have indicated that perception during binocular rivalry does not result solely from reciprocal inhibitory competition between monocular channels. In this study we were interested in feature selectivity both during dominance and during suppression. The first experiment shows that a suppressed stimulus perceptually appears earlier when it shares features with a visible stimulus than when it does not. Subsequently, our second experiment suggests a reversal of this effect when similarity is exhibited with a suppressed stimulus. These findings hint at a role for both selective enhancing (Experiment 1) and selective inhibitory cortical mechanisms (Experiment 2) in causing image rivalry. From a phenomenological perspective these results suggest that we are not only selectively aware but also selectively unaware of specific features in the visual scene.     

Both monocular and binocular signals contribute to motion rivalry

There is an ongoing debate on whether binocular rivalry involves competition among monocular cells or binocular cells. We investigated this issue psychophysically with two specially designed test stimuli. One test stimulus contained monocular motion signals but greatly reduced binocular motion signals, while the other contained binocular motion signals but no monocular motion signals. For comparison, we also employed a normal rivalrous control containing both monocular and binocular motion signals, and a non-rivalrous flicker-noise control with neither monocular nor binocular motion signals. We found that binocular rivalry for the two test stimuli was significantly reduced compared with the normal rivalrous control, but not completely eliminated compared with the non-rivalrous control. Therefore, both monocular and binocular motion signals appear to contribute to motion rivalry, suggesting that motion rivalry must involve competition among both monocular and binocular cells.     

Center-surround inhibition deepens binocular rivalry suppression

When dissimilar stimuli are presented to each eye, perception alternates between both images--a phenomenon known as binocular rivalry. It has been shown that stimuli presented in proximity of rival targets modulate the time each target is perceptually dominant. For example, presenting motion to the region surrounding the rival targets decreases the predominance of the same-direction target. Here, using a stationary concentric grating rivaling with a drifting grating, we show that a drifting surround grating also increases the depth of binocular rivalry suppression, as measured by sensitivity to a speed discrimination probe on the rival grating. This was especially so when the surround moved in the same direction as the grating, and was slightly weaker for opposed directions. Suppression in both cases was deeper than a no-surround control condition. We hypothesize that surround suppression often observed in area MT (V5)-a visual area implicated in visual motion perception-is responsible for this increase in suppression. In support of this hypothesis, monocular and binocular surrounds were both effective in increasing suppression depth, as were surrounds contralateral to the probed eye. Static and orthogonal motion surrounds failed to add to the depth of rivalry suppression. These results implicate a higher-level, fully binocular area whose surround inhibition provides an additional source of suppression which sums with rivalry suppression to effectively deepen suppression of an unseen rival target.     

Strength and coherence of binocular rivalry depends on shared stimulus complexity

Presenting incompatible images to the eyes results in alternations of conscious perception, a phenomenon known as binocular rivalry. We examined rivalry using either simple stimuli (oriented gratings) or coherent visual objects (faces, houses etc). Two rivalry characteristics were measured: Depth of rivalry suppression and coherence of alternations. Rivalry between coherent visual objects exhibits deep suppression and coherent rivalry, whereas rivalry between gratings exhibits shallow suppression and piecemeal rivalry. Interestingly, rivalry between a simple and a complex stimulus displays the same characteristics (shallow and piecemeal) as rivalry between two simple stimuli. Thus, complex stimuli fail to rival globally unless the fellow stimulus is also global. We also conducted a face adaptation experiment. Adaptation to rivaling faces improved subsequent face discrimination (as expected), but adaptation to a rivaling face/grating pair did not. To explain this, we suggest rivalry must be an early and local process (at least initially), instigated by the failure of binocular fusion, which can then become globally organized by feedback from higher-level areas when both rivalry stimuli are global, so that rivalry tends to oscillate coherently. These globally assembled images then flow through object processing areas, with the dominant image gaining in relative strength in a form of 'biased competition', therefore accounting for the deeper suppression of global images. In contrast, when only one eye receives a global image, local piecemeal suppression from the fellow eye overrides the organizing effects of global feedback to prevent coherent image formation. This indicates the primacy of local over global processes in rivalry.     

Grouping visual features during binocular rivalry

During binocular rivalry, portions of one eye's view may be perceptually dominant while other portions are suppressed; at any given moment, overall dominance often resembles a patchwork mixture of the two eyes' views. This study investigates the potency of two Gestalt grouping cues--good continuation and common fate--to promote synchronous fluctuations in dominance of two, spatially separated rival targets. Two grating patches were presented to the left eye paired dichoptically with random-dot patches presented to corresponding right eye locations. The orientations of the two gratings were either collinear, parallel or orthogonal. Gratings underwent contrast modulations that were either correlated (identical contrast changes) or uncorrelated (independent contrast changes). Over 60 s trials, observers pressed one key when the left grating predominated, another when the right grating predominated and both keys when both were concurrently visible. Correlated contrast modulation promoted joint grating predominance relative to the uncorrelated conditions, an effect strongest for collinear gratings. Joint predominance depended strongly on the angular separation between gratings and the temporal phase-lag in contrast modulations. These findings may reflect neural interactions subserved by lateral connections between cortical hypercolumns.     

Center-surround interactions in visual motion processing during binocular rivalry

When each eye is confronted with a dissimilar stimulus, the percept will generally alternate between the two. This phenomenon is known as binocular rivalry. Although binocular rivalry occurs at locations where targets overlap spatially, the area surrounding rivalrous targets can modulate their dominance. Here we show that during binocular rivalry of oppositely moving gratings, a surrounding grating moving in the same direction as one of the two leads to increased dominance of the opposite direction of motion in the center. This increased dominance of the opposite direction in the center was observed irrespective of the eye to which the surround was presented. Inspection of the results for different conditions reveals that the preference for the opposite direction of motion cannot be explained by a single mechanism operating beyond binocular fusion. We therefore suggest that this phenomenon is the outcome of center-surround interactions at multiple levels along the pathway of visual motion processing.     

On the relation between dichoptic masking and binocular rivalry

When our two eyes view incompatible images, the brain invokes suppressive processes to inhibit one image, and favor the other. Two phenomena are typically observed: dichoptic masking (reduced sensitivity to one image) for brief presentations, and binocular rivalry (alternation between the two images), over longer exposures. However, it is not clear if these two phenomena arise from a common suppressive process. We investigated this by measuring both threshold elevation in simultaneous dichoptic masking and mean percept durations in rivalry, whilst varying relative stimulus orientation. Masking and rivalry showed significant correlations, such that strong masking was associated with long dominance durations. A second experiment suggested that individual differences across both measures are also correlated. These findings are consistent with varying the magnitude of interocular suppression in computational models of both rivalry and masking, and imply the existence of a common suppressive process. Since dichoptic masking has been localised to the monocular neurons of V1, this is a plausible first stage of binocular rivalry.     

A bias for looming stimuli to predominate in binocular rivalry

Concentric gratings that expand outwards are seen for a greater period of time relative to contracting gratings when engaged in binocular rivalry. During binocular rivalry (BR), which is a fluctuation in visual awareness between different images presented separately to each eye, equivalent images tend to be seen in equal proportion over the observation period. When one eye's image is particularly salient, brighter, or moving, this equality is curtailed, and the stronger image predominates. Here a specific direction of motion is found to predominate over another of equal speed. This tendency is consistent with the ability of looming objects to orient attention, coupled with previous accounts of the role of stimulus-driven attention in BR.     

Adaptive center-surround interactions in human vision revealed during binocular rivalry

We used binocular rivalry as a psychophysical probe to explore center-surround interactions in orientation, motion and color processing. Addition of the surround matching one of the rival targets dramatically altered rivalry dynamics. For all visual sub-modalities tested, predominance of the high-contrast rival target matched to the surround was greatly reduced-a result that disappeared at low contrast. At low contrast, addition of the surround boosted dominance of orientation and motion targets matched to the surround. This contrast-dependent modulation of center-surround interactions seems to be a general property of the visual system and may reflect an adaptive balance between surround suppression and spatial summation.     

A binocular rivalry study of motion perception in the human brain

The relationship between brain activity and conscious visual experience is central to our understanding of the neural mechanisms underlying perception. Binocular rivalry, where monocular stimuli compete for perceptual dominance, has been previously used to dissociate the constant stimulus from the varying percept. We report here fMRI results from humans experiencing binocular rivalry under a dichoptic stimulation paradigm that consisted of two drifting random dot patterns with different motion coherence. Each pattern had also a different color, which both enhanced rivalry and was used for reporting which of the two patterns was visible at each time. As the perception of the subjects alternated between coherent motion and motion noise, we examined the effect that these alternations had on the strength of the MR signal throughout the brain. Our results demonstrate that motion perception is able to modulate the activity of several of the visual areas which are known to be involved in motion processing. More specifically, in addition to area V5 which showed the strongest modulation, a higher activity during the perception of motion than during the perception of noise was also clearly observed in areas V3A and LOC, and less so in area V3. In previous studies, these areas had been selectively activated by motion stimuli but whether their activity reflects motion perception or not remained unclear; here we show that they are involved in motion perception as well. The present findings therefore suggest a lack of a clear distinction between 'processing' versus 'perceptual' areas in the brain, but rather that the areas involved in the processing of a specific visual attribute are also part of the neuronal network that is collectively responsible for its perceptual representation.     

An integrative model of binocular vision: a stereo model utilizing interocularly unpaired points produces both depth and binocular rivalry

Half-occluded points (visible only in one eye) are perceived at a certain depth behind the occluding surface without binocular rivalry, even though no disparity is defined at such points. Here we propose a stereo model that reconstructs 3D structures not only from disparity information of interocularly paired points but also from unpaired points. Starting with an array of depth detection cells, we introduce cells that detect unpaired points visible only in the left eye or the right eye (left and right unpaired point detection cells). They interact cooperatively with each other based on optogeometrical constraints (such as uniqueness, cohesiveness, occlusion) to recover the depth and the border of 3D objects. Since it is contradictory for monocularly visible regions to be visible in both eyes, we introduce mutual inhibition between left and right unpaired point detection cells. When input images satisfy occlusion geometry, the model outputs the depth of unpaired points properly. An interesting finding is that when we input two unmatched images, the model shows an unstable output that alternates between interpretations of monocularly visible regions for the left and the right eyes, thereby reproducing binocular rivalry. The results suggest that binocular rivalry arises from the erroneous output of a stereo mechanism that estimates the depth of half-occluded unpaired points. In this sense, our model integrates stereopsis and binocular rivalry, which are usually treated separately, into a single framework of binocular vision. There are two general theories for what the "rivals" are during binocular rivalry: the two eyes, or representations of two stimulus patterns. We propose a new hypothesis that bridges these two conflicting hypotheses: interocular inhibition between representations of monocularly visible regions causes binocular rivalry. Unlike the traditional eye theory, the level of the interocular inhibition introduced here is after binocular convergence at the stage solving the correspondence problem, and thus open to pattern-specific mechanisms.     

Perceptual dominance during binocular rivalry is prolonged by a dynamic surround

We examined whether dynamic stimulation that surrounds a rival target influences perceptual alternations during binocular rivalry. We presented a rival target surrounded by dynamic random-dot patterns to both eyes, and measured dominance durations for each eye’s rival target. We found that rival target dominance durations were longer when surrounds were dynamic than when they were static or absent. Additionally, prolonged dominance durations were more apparent when the dynamic surround was alternately presented between the two eyes than when it was presented simultaneously to both eyes. These results indicate that dynamic stimulation that surrounds a rival target plays a role in maintaining the current perceptual state, and causes less perceptual alternations during binocular rivalry. Our findings suggest that dynamic signals on the retina may suppress rivalry, and thus provide useful information for stabilizing perceptions in daily life.     

Contrast sensitivity of form and motion discrimination during binocular rivalry

Binocular rivalry, which is induced by presenting the two eyes with incompatible stimuli, results in periods where one eye's stimulus is seen and the other stimulus is suppressed. We measured the depth of suppression in two ways, with very different results. First, two similar forms were briefly presented to one eye: the difference in shapes required to discriminate the forms was substantially greater during suppression than during dominance. Second, the two forms were made sufficiently different in shape to be easily distinguishable at high contrast, and contrast was lowered to find the threshold for discrimination of the forms. Contrast sensitivity did not differ between the suppression and dominance states. These results were replicated with a motion discrimination task: suppression markedly worsened the ability to distinguish increases from decreases in speed but did not elevate the minimum contrast required for the same task. We interpret the results in terms of steep contrast-response functions in visual cortex beyond the primary area.     

Detecting contrast changes in invisible patterns during binocular rivalry

When dissimilar images are presented to the two eyes, the human visual system lapses into binocular rivalry, a unique perceptual state characterized by stochastic alternations in dominance of one of the two source images over the other. Probe targets delivered to an eye during suppression phases are more difficult to detect than probes delivered during dominance phases. Nearly all probe studies have involved presenting new stimulation (e.g., a spot of light) either superimposed on or replacing the suppressed stimulus. Here, we ask whether observers can detect a reduction in the contrast of the suppressed stimulus itself. In other words, can observers detect a probe that should make an already invisible stimulus even weaker? Specifically, we compared observers’ ability to detect contrast increments and contrast decrements introduced within a rival pattern during dominance and suppression. Contrast increment thresholds were elevated across all pedestal contrasts when the increment was introduced during suppression compared to during dominance, replicating previous results. Contrast decrement thresholds measured during suppression were elevated to an even greater extent, but the fact that they were obtained at all establishes that observers were able to detect probes that should make an already invisible target even more difficult to perceive. In a second experiment, we found a similar pattern of results for contrast change detection in complex images of faces as well. Based on the resulting threshold-vs.-contrast functions, we suggest that, regardless of the complexity of the image, rivalry suppression modulates the neural contrast response function through a mixture of reduced overall response gain and a shift in the contrast gain.     

Caloric vestibular stimulation reveals discrete neural mechanisms for coherence rivalry and eye rivalry: a meta-rivalry model

Binocular rivalry is an extraordinary visual phenomenon that has engaged investigators for centuries. Since its first report, there has been vigorous debate over how the brain achieves the perceptual alternations that occur when conflicting images are presented simultaneously, one to each eye. Opposing high-level/stimulus-representation models and low-level/eye-based models have been proposed to explain the phenomenon, recently merging into an amalgam view. Here, we provide evidence that during viewing of Díaz-Caneja stimuli, coherence rivalry -- in which aspects of each eye's presented image are perceptually regrouped into rivalling coherent images -- and eye rivalry operate via discrete neural mechanisms. We demonstrate that high-level brain activation by unilateral caloric vestibular stimulation shifts the predominance of perceived coherent images (coherence rivalry) but not half-field images (eye rivalry). This finding suggests that coherence rivalry (like conventional rivalry according to our previous studies) is mediated by interhemispheric switching at a high level, while eye rivalry is mediated by intrahemispheric mechanisms, most likely at a low level. Based on the present data, we further propose that Díaz-Caneja stimuli induce 'meta-rivalry' whereby the discrete high- and low-level competitive processes themselves rival for visual consciousness.