|Randolph Blake and Frank Tong (2008), Scholarpedia, 3(12):1578.||doi:10.4249/scholarpedia.1578||revision #124436 [link to/cite this article]|
Binocular rivalry is a visual phenomenon that occurs when dissimilar monocular stimuli are presented to corresponding retinal locations of the two eyes. Rather than perceiving a stable, single amalgam of the two stimuli, one experiences alternations in perceptual awareness over time as the two stimuli compete for perceptual dominance. Binocular rivalry is a compelling example of multistable perception wherein physically invariant stimulation leads to fluctuations in perception. It is also one of a few psychophysical phenomena that have been usefully exploited to study visual processing outside of awareness; other such phenomena include metacontrast masking, motion induced blindness, and flash suppression.
People with normal binocular vision do not usually experience rivalry alternations under ordinary viewing conditions, even though many local regions in each eye may undergo suppression. Alternations in rivalry dominance usually occur after a few seconds of steadying viewing, but during natural viewing, the eyes rarely remain fixed for more than a few hundred milliseconds. Nonetheless, the neural processes underlying rivalry may still be operating to promote single vision by eliminating double vision (diplopia) and local interocular conflicts concerning the positions of objects in 3D visual space. Most believe that binocular rivalry, along with other forms of perceptual multistability, reflects neural processes involved in resolving visual conflict arising from inherently ambiguous sensory signals (e.g., Leopold & Logothetis, 1999). A few think rivalry reveals the operation of an intrinsic neural oscillator controlling a host of rhythmic behaviors (Pettigrew, 2001). In recent years, neuroscientists have adopted binocular rivalry as a means for identifying neural concomitants of visual awareness. Using single-cell recording techniques in nonhuman primates and brain imaging techniques in humans, neuroscientists have identified time-varying changes in neural responses correlated with alternations in perceptual dominance during rivalry.
Portions of this essay are adapted from a review of binocular rivalry appearing in the journal Mind and Brain, which is no longer in print (Blake, 2001).
According to noted vision scholar Nicholas Wade (1998), it is Giambattista della Porta who deserves credit for the ﬁrst unambiguous description of rivalry. Writing in the 16th century, Porta described viewing different pages from two books, one page with each eye, and being able to read one of the two pages without interference from the other. It was Étienne-Francois Dutour, writing in the 18th century, who provided the first vivid description of the alternations in perceptual dominance when the two eyes view dissimilar stimulation. But it is Sir Charles Wheatstone who deserves credit for the ﬁrst systematic study of rivalry, contained in his famous 19th century monograph on binocular stereopsis. Using his newly invented mirror stereoscope ( Figure 1a), Wheatstone observed a fascinating disruption of stable binocular vision when the two eyes separately viewed different alphabetic letters. To quote Wheatstone:
- If a and b [shown as S and A in Figure 1b] are each presented at the same time to a different eye, the common border will remain constant, while the letter within it will change alternately from that which would be perceived by the right eye alone to that which would be perceived by the left eye alone. At the moment of change the letter which has just been seen breaks into fragments, while fragments of the letter which is about to appear mingle with them, and are immediately after replaced by the entire letter. It does not appear to be in the power of the will to determine the appearance of either of the letters, but the duration of the appearance seems to depend on causes which are under our control: thus if the two pictures be equally illuminated, the alternations appear in general of equal duration; but if one picture be more illuminated than the other, that which is less so will be perceived during a shorter time.
In this short passage Wheatstone identifies several key aspects of rivalry, including suppression of one of two discordant stimuli, alternations in dominance between the eyes, spatial fragmentation of the two images during times of transition, and the influence of relative stimulus strength on the predominance of one rival stimulus over the other.
Following Wheatstone’s seminal essay, binocular rivalry captured the imagination of some of science’s leading ﬁgures, including German physicist/physiologist Hermann von Helmholtz, American psychologist William James and English physiologist Sir Charles Sherrington. During the 20th century literally hundreds of papers were published on the topic, with interest in binocular rivalry accelerating toward the end of the century. Papers on binocular rivalry published during the middle decades of the 20th century were reviewed by Levelt (1965) and by Walker (1978), and a comprehensive bibliography of those early papers can be found here.
More contemporary work has been reviewed by Blake and Logothetis (2002), and an edited volume with chapters devoted entirely to binocular rivalry appeared in 2005 (Alais & Blake, 2005).
What instigates binocular rivalry?
Left- and right-eye image differences along any one of a wide range of stimulus dimensions are sufficient to instigate binocular rivalry ( Figure 2). These include interocular differences in color, luminance, contrast polarity, form, size, or velocity. Rivalry can be triggered by very simple stimulus differences (e.g., gratings differing only in orientation) and by differences between complex images (e.g., pictures of a human face viewed by one eye and of a house viewed by the other). Stronger, high-contrast stimuli lead to stronger perceptual competition, as indicated by faster rivalry alternations. Interestingly though, even a faint low-contrast pattern that is barely visible will eventually suppress a strong one. Rivalry can even occur under dim (scotopic) viewing conditions, when light levels are so low they can only be detected by the retina's rod photoreceptors. Rivalry can be observed anywhere throughout the binocular visual ﬁeld so long as the rival stimuli are adjusted for visibility. Under some conditions, rivalry can be triggered by physically identical stimuli that differ in appearance owing to simultaneous luminance or color contrast. Rivalry cannot be triggered, however, by global form differences between left- and right-eye images when the local elements comprising those forms are compatible between the eyes.
Not all left- and right-eye image differences trigger rivalry; the following do not rival but instead meld into either a superimposition of the two monocular views (binocular transparency) or a stable average of the two (binocular fusion). Rivalry does not occur when the two eyes view uncontoured patches of light flickering at different rates to the two eyes. Nor is rivalry experienced when dichoptically viewing two gratings differing by several octaves or more in spatial frequency or when dichoptically viewing two sets of moving dots differing greatly in speed; these failures of rivalry suggest that interocular competition occurs within circumscribed "visual channels" comprising neurons with similar spatial or spatiotemporal tuning properties. When two monocular images differ in luminance or in contrast but share a common polarity, the binocular impression takes on the appearance of the fused average of two monocular components. When dichoptic orthogonal gratings are presented at very low contrast levels, just above the threshold required for visibility, the two weakly conflicting images will appear as a stable fused plaid for many seconds before rivalry finally ensues.
Alternations in perceptual dominance – the hallmark of binocular rivalry – can also be produced by rapidly and repetitively swapping dissimilar monocular stimuli between the two eyes. With this form of perceptual bistability, dubbed stimulus rivalry, periods of perceptual dominance transcend multiple eye swaps, so the resulting perceptual experience cannot be attributed to competition between the eyes. Fluctuations in perceptual dominance can also be experienced when one or both eyes view dissimilar patterns that are physically superimposed. This form of perceptual bistability, dubbed monocular rivalry, also rules out eye competition. Compared to conventional binocular rivalry, these other forms of perceptual rivalry tend to occur under more restricted stimulus conditions, and they involve fewer periods of exclusive dominance of one stimulus or the other.
Fluctuations in perceptual interpretation are also experienced when viewing any of the well-known ambiguous ﬁgures, such as the Necker cube and the Rubin vase/face figure; these forms of perceptual bistability arise from competing perspective interpretations or from incompatible figure/ground interpretations. Bistable perception can arise, too, when viewing animation sequences portraying ambiguous 3D structure, conflicting paths of motion, or alternative grouping interpretations. These forms of bistable perception involve alternations in the grouping of visual elements that remain continuously visible, whereas during binocular rivalry it is the visual elements that alternately disappear from view. Finally, bistable auditory perception can be produced when listening to tone sequences that can be heard either as a single, grouped auditory stimulus or as two independent streams of sounds; examples of auditory bistability can be heard at this website.
Spatial properties of binocular rivalry
When viewing relatively large rival stimuli, one often experiences ﬂuctuating patchworks of dominance consisting of intermingled portions of both eyes’ views; this is termed piecemeal rivalry in the literature. Evidently this dependence of piecemeal rivalry on rival target size is governed by retinal image size, not perceived size, because coherent, non-piecemeal rivalry is experienced when small rival images appear large in virtue of Emmert's law. Periods of mixed dominance imply that rivalry transpires locally within spatially restricted “zones” and not globally between the eyes. How large are these supposed zones? For foveally viewed rival stimuli, it has been estimated that exclusive, unitary rivalry with no periods of mixture can be obtained only when rival stimuli are smaller than 0.5 deg visual angle. However, coherent rivalry can occur with larger targets if they are presented at lower spatial frequencies, placed further in periphery, or are presented under dim light (scotopic conditions). Suppression can also spread some distance beyond the immediate regions of binocular conﬂict, also implying the existence of zones of rivalry. These zones of rivalry become increasingly larger as a function of visual distance from fixation (i.e., eccentricity), in a manner that matches the cortical magnification properties of the primary visual cortex.
Although coherent rivalry may be found only within a small restricted zone of the visual field, many observations suggest that some forms of rivalry interactions can occur across neighboring zones. First, transitions in dominance from one stimulus to the other often appear like a wave: one stimulus appears to sweep over the other, erasing the latter from visual awareness. This implies that the zones of rivalry are interconnected. Second, even a relatively large stimulus, while susceptible to piecemeal rivalry, will achieve complete dominance signiﬁcantly more often than would be expected based on independent zones. Third, multiple small rival targets scattered throughout the visual ﬁeld can engage in synchronized alternations, such that all targets of a given conﬁguration are dominant simultaneously (see, for example, Figure 2b). Fourth, visual features located outside the boundaries of a rival target can inﬂuence the predominance of that target, implying that the target’s strength is being modulated by its surrounding context. Thus, rivalry includes local competitive interactions, leading to exclusive dominance within a region or zone, as well as more global interactions that can facilitate the spread of dominance across large portions of the visual field.
Temporal properties of binocular rivalry
Binocular rivalry fails to occur when dissimilar monocular targets are presented for durations less than 200 milliseconds – with very brief presentations, one instead experiences the binocular superimposition of the two targets. So, for example, a pair of brieﬂy ﬂashed, orthogonally oriented rival gratings looks like a plaid. To experience complete dominance of one rival ﬁgure over another requires that both rival ﬁgures be presented simultaneously for at least several hundred milliseconds. During rivalry, the currently suppressed stimulus can be readily restored to dominance by any maneuver that creates transient stimulation within the suppressed stimulus. For example, simply waving a pencil in front of a suppressed stimulus almost immediately brings it into dominance, erasing the other eye’s view from consciousness. Reversals in dominance can also be achieved by abruptly increasing the contrast of a suppressed stimulus or by setting a suppressed stimulus into motion. On the other hand, perceptual dominance can be stabilized over time to favor one rival stimulus if the rivalry display is repeatedly shown using brief, intermittent presentation (Pearson & Brascamp, 2008).
When rival targets are viewed for an extended period of time without perturbation to either stimulus, the individual periods of dominance and suppression are unpredictable in duration, and collectively these durations conform reasonably well to a unimodal distribution skewed toward longer durations ( Figure 3). The rate of alternations in dominance, defined as the number of reversals in dominance state per unit time, increases with the strength and complexity of the rival stimuli. For example, low contrast rival stimuli yield longer dominance durations and, hence, slower alternation rates than do high contrast stimuli. Rivalry rate also varies significantly among individuals, and alternation rate generally slows with age. It may be that these individual differences arise, at least in part, from differing patterns of eye movements which, of course, shift the retinal positions of the image of rival stimuli: saccadic eye movements can both trigger a change in rivalry state or, alternatively, can prolong the dominance of the currently dominant stimulus, depending on the timing and the extent of those saccades. Ongoing rivalry alternations can also be perturbed by transcranial magnetic stimulation (TMS) that targets the occipital pole and, therefore, visual areas V1 and V2.
A widely used index of rivalry dynamics is predominance, the total percentage of viewing time that a given stimulus is dominant. Another important measure is dominance duration ( Figure 3), reflecting how long a given stimulus can maintain exclusive dominance, until its competitor finally breaks suppression. For left- and right-eye rival stimuli matched in strength and complexity, predominance tends to be equal, whereas predominance tips in favor of one stimulus simply by varying unilaterally the strength of one stimulus relative to the other (Levelt, 1965). Manipulation of stimulus variables such as luminance, contrast, contour density, spatial frequency, size, velocity, and visual field location can produce pronounced variations in alternation rate and in stimulus predominance. As a rule, a “stronger” rival target (e.g., one that is higher contrast than the other) enjoys enhanced predominance primarily by suppressing its competitor for longer periods of time. Increasing the strength of a rival target can sometimes weakly increase that target's dominance duration, but has a much greater influence in reducing the dominance duration of the competing target. This asymmetrical effect is predicted by reciprocal inhibition models of binocular rivalry.
It is widely believed that transitions in dominance from one stimulus to the other are mediated, at least in part, by neural adaptation, with the excitatory signals from the dominant stimulus weakening over time and eventually succumbing to the signals from the other, previously suppressed stimulus. Consistent with this idea, a stimulus forced to remain dominant during rivalry (by incrementing its contrast the moment it becomes suppressed) generates increasingly brief dominance durations, as if being denied an opportunity to recover from adaptation; once dominance is no longer forced, that stimulus remains suppressed for an unusually long period of time. Conversely, abnormally long dominance durations are produced when rival targets move around the visual field in a manner that precludes neural adaptation of a given region of the visual field. Neural adaptation alone, however, might not provide a complete account of rivalry alternations, because many recent models of rivalry invoke neural noise as a causal agent in triggering switches in perceptual state. Noise could result from small eye movements and microsaccades leading to small image displacements on the retina, as well as physiological fluctuations in the central nervous system.
Top-down influences on rivalry perception
When viewing rival stimuli, observers seem unable willfully to trigger immediate switches from suppression to dominance or to hold one stimulus dominant indeﬁnitely. One can deploy attention, however, to bias perception in favor of one of two rival ﬁgures during extended viewing periods. Helmholtz observed that he could bias dominance in favor of one rival target by concentrating his attention on that target using strategies like counting the number of lines in a given rival stimulus. Even then, however, Helmholtz found that attention’s efﬁcacy seemed to wane, for a dominant stimulus eventually gives way to suppression involuntarily after a period of time. Helmholtz’s observations have since been documented in a number of studies showing that people can use attention to bias initial dominance at the onset of rival stimulation and can inﬂuence the alternation rate of ongoing rivalry. Psychophysical studies suggest attention may enhance the dominance of a rival stimulus partly by increasing its effective contrast, as attentional effects can be simulated by increasing the contrast of a rival target during periods of dominance.
Rivalry dynamics can also be influenced by cognitive and motivational factors. In one older study, for example, Jewish and Catholic observers judged the relative predominance of symbols associated with their two religions (the star of David versus a cross). Jewish religious symbols tended to predominate for the Jewish observers whereas Catholic religious symbols tended to predominate for the Catholic observers. In a similar vein, upright pictures of a human face tend to predominate over inverted pictures of a face, and fearful faces tend to predominate over neutral ones. Gaze direction, too, influences rivalry, with direct eye gaze being more salient than averted eye gaze. Even mental imagery of a pattern can facilitate the subsequent likelihood of that pattern’s predominating during binocular rivalry, with longer periods of imagery leading to stronger top-down bias effects. Several recently published studies even show that rivalry dynamics are susceptible to influences from odors, sounds and motor responses. Results such as these imply that the resolution of stimulus conﬂict during binocular rivalry is influenced by a host of sensory, cognitive, and affective factors, implying the involvement of multiple, interconnected brain areas in the control of binocular rivalry (Tong et al., 2006).
What rivals during binocular rivalry?
When viewing dissimilar monocular stimuli, one experiences alternations in perceptual dominance between those stimuli, seeing either one or the other for several seconds at a time with no sense whatsoever about which eye is dominant. From a phenomenological standpoint, in other words, rivalry appears to involve competition between alternative stimulus interpretations, not rivalry between corresponding regions of the eyes. Yet it is clear that the visual system both retains and relies on eye-of-origin information, regarding which eye is receiving which rival stimulus and, therefore, which eye is currently dominant. For example, when dominant and suppressed images are swapped between the eyes in a manner precluding disturbing transients, observers will see the previously suppressed stimulus that has now been introduced to the previously dominant eye -- they do not continue to see the previously dominant stimulus. Moreover, all sorts of probe stimuli presented to an eye during suppression are more difficult to detect, regardless of their similarity to the currently suppressed stimulus. For that matter, ordinarily noticeable changes in a rival stimulus itself (e.g., reversal in the direction of motion of an array of dots) go unnoticed for several seconds when those changes occur during suppression. These observations imply that it is not a given stimulus that is suppressed but, rather, a region of the eye.
Yet other findings point to stimuli, not eyes, as the agents of rivalry. For example, rival targets differing in form and color can sometimes achieve states of dominance in which the color from one eye’s image combines in dominance with the form from the other eye’s image, temporarily producing a binocular impression corresponding to neither eye’s stimulus. Likewise, perceived direction of motion of a dominant stimulus can be inﬂuenced by the velocity of a suppressed stimulus. Results such as these show that perceptual experience during dominance phases of rivalry sometimes comprise an amalgam of attributes from both eyes’ views, which should be impossible if all information from a given region of one eye were lost during suppression. Also cited in support of stimulus-based rivalry is interocular grouping during rivalry ( Figure 4 and also Figure 2b), in which very speciﬁc combinations of left-eye and right-eye components achieve dominance simultaneously to form a coherent perceptual experience. Obviously, inputs from both eyes must be contributing to dominance in these instances. Such interocular grouping could occur at a binocular level of representation, after inputs from the two eyes have been combined, and also at the monocular level, with facilitatory interactions occurring between left-eye and right-eye neurons with similar feature preferences.
The controversy over what rivals during rivalry - eye versus stimulus - has largely been resolved in favor of a hybrid model embodying both forms of competition transpiring at multiple levels within the visual pathways (Blake & Logothetis, 2002; Tong et al., 2006).
Visual processing outside of awareness: what survives interocular suppression?
Binocular rivalry affords a unique opportunity to discover aspects of perceptual processing that transpire outside of visual awareness. Over the years, a number of studies have sought to determine whether some visual processes can proceed unimpeded, even when the stimuli mediating those processes are rendered invisible by rivalry suppression. It is known, for instance, that several classic adaptation aftereffects can be generated, albeit more weakly, by stimuli rendered invisible by binocular rivalry suppression; these include the motion aftereffect, the tilt aftereffect, and the McCollough effect. On the other hand, other adaptation aftereffects are completely abolished by suppression of the inducing stimuli, and these include adaptation to global motion and to faces. Suppression of a stimulus also weakens or abolishes that stimulus’s ability to promote visual priming (i.e., to increase the speed and accuracy of identifying subsequently viewed words or pictures). In recent years, this strategy of measuring stimulus effectiveness during suppression has been extended to brain imaging studies, with at least some results showing reliable neural responses generated to certain classes of stimuli rendered invisible by interocular suppression (Fang & He, 2005).
Binocular rivalry and stereopsis
What about binocular rivalry’s effect on stereopsis? At first glance, the two phenomena seem incompatible. Stereopsis represents a form of interocular cooperation in which the brain registers slight perspective differences between those views to create a stable, three-dimensional representation incorporating both views. Rivalry, in contrast, involves competition between stimuli presented to the two eyes. Is it possible for rivalry and stereopsis to coexist at a common visual location? The answer seems to be a qualified “yes”. For example, color rivalry and stereopsis can co-occur if a common stimulus is shown to both eyes (with one slightly displaced), and those two monocular stimuli differ only in color but share a common luminance polarity. Common luminance information allows for fusion and stereopsis, while the discrepant color component still allows for rivalry. However, if the two shapes differ in luminance polarity, stereo-depth perception is greatly impaired and the observer will experience rivalry or binocular lustre. A more revealing experiment involves the simultaneous presentation of stereoscopic stimuli to the two eyes and a rival stimulus superimposed on one of those stereo images in one eye. Can neural processing supporting stereopsis co-exist in the face of ongoing rivalry between disparate forms? Here, one can measure stereoscopic sensitivity under conditions of rivalry, by continuously presenting one stereo half-image to one eye and a rival target to the other eye. Once the stereo half-image undergoes suppression (as indicated by an observer’s button press), the other stereo half-image can be superimposed on the dominant rival target for varying durations. If the half-image is flashed for only a few hundred milliseconds, stereopsis is impaired, but stereo sensitivity improves with longer durations and eventually reaches normal levels. Interestingly, once stereofusion is achieved, these stereo components can no longer effectively rival with the unpaired "rival" half-image. Thus, stereopsis overrides rivalry, after a brief initial period during which rivalry dominates. The broader implication is that rivalry represents a default outcome when the brain fails to register matching features in the left and right eyes’ images.
Neural bases of binocular rivalry
It is now generally recognized that binocular rivalry does not occur at one particular neural locus. Brain imaging studies of humans experiencing rivalry (Tong et al., 2006) and single-unit studies from awake, behaving monkeys experiencing rivalry (Logothetis, 1998) both imply that traces of the neural signature of rivalry are detectable at early stages of visual processing, with those traces then being ampliﬁed at subsequent, higher stages. Both neurophysiological and human neuroimaging studies have found very powerful modulations in high-level object-sensitive regions of the ventral temporal cortex that reflect the observer’s awareness during rivalry. However, in early visual areas such as V1, there have been differences in reports of the strength of rivalry modulations. In general, neuroimaging measures of BOLD activity have found strong modulations in V1, and have even detected modulations in the LGN. In comparison, electrophysiological recordings in area V1 of the monkey have found somewhat weaker modulations in the local field potential and considerably weaker modulations in single-unit activity. It is possible that much of the modulatory activity in V1 is presynaptic, perhaps due to suppressive effects of feedback targeting V1. Another possibility is that rivalry involves strong modulations of synaptic and spiking activity for a subset of neurons, but that for some reason these neurons are difficult to isolate using standard recording methods, so their activity can only be detected by relying on measures sensitive to population activity. Regardless, current evidence strongly suggests that rivalry is the culmination of a cascade of neural events transpiring at multiple sites along the visual pathways, with no single site constituting “the” locus of rivalry.
As described in earlier sections, binocular rivalry can be characterized in terms of its spatial extent, its temporal dynamics, and its generality beyond those conditions triggering rivalry. These aspects of rivalry are not necessarily tied to a single, omnibus process. The stimulus determinants of switches between dominance and suppression (i.e., rivalry's temporal dynamics) are not necessarily the same as those governing the spatial extent of rivalry. Similarly, the stimulus properties influencing the durations of suppression phases (e.g., luminance and contrast) differ from those properties controlling dominance durations (e.g., global context).
When considering the neural bases of rivalry, it is important to distinguish between processes responsible for initiating rivalry and for selecting one eye’s input for dominance from processes responsible for implementing and maintaining suppression. Suppression and selection probably result from separate mechanisms operating at different stages in the visual system. In a sense, selection involves the comparison of information presented to the two eyes to establish the degree of correspondence and, perhaps, the relation of a given stimulus to other features within the visual ﬁeld. Once selection has been accomplished, one stimulus (or portions of it) is temporarily “rejected” (i.e., suppressed) while another stimulus dominates. The fate of the suppressed stimulus – and other, new information presented within its immediate vicinity – can be determined by neural events different from those registering the initial incompatibility of monocular stimuli. This idea of rivalry resulting from a cascade of distributed neural processes can reconcile seemingly conﬂicting results within the rivalry literature (Tong et al., 2006).
- Alais, D. & Blake, R. (2005) Binocular rivalry and perceptual ambiguity, MIT Press, Cambridge MA.
- Blake, R. & N. K. Logothetis (2002). "Visual Competition." Nature Reviews: Neuroscience 3(1): 13-21.
- Blake, R. (2001). "A Primer on binocular rivalry, including current controversies." Brain and Mind 2: 5-38.
- Brascamp, J.W., van Ee, R., Pestman, W. and van den Berg, A.V. (2005) Distributions of alternation rates in various forms of bistable perception. Journal of Vision, 5(4), 287-298.http://journalofvision.org/5/4/1/
- Fang, F. & He, S. (2005). "Cortical responses to invisible objects in the human dorsal and ventral pathways." Nature Neuroscience 8(10): 1380-1385.
- Kovács, I. et al. (1996). "When the brain changes its mind: Interocular grouping during binocular rivalry." Proceedings of the National Academy of Sciences of the United States of America 93: 15508-15511.
- Leopold, D.A. & Logothetis, N.K. (1999) “Multistable phenomena: changing views in perception.” Trends in Cognitive Science 3: 254-264.[]
- Levelt, W. J. M. (1965). On binocular rivalry, Institute for Perception RVO-TNO, Soesterberg, The Netherlands
- Logothetis, N. (1998). "Single units and conscious vision." Philosophical Transactions of the Royal Society, London B Biological Sciences 353: 1801-1818.[]
- Pearson, J. & Brascamp, J. (2008) Sensory memory for ambiguous vision. Trends in Cognitive Sciences, 12: 334-341.
- Pettigrew, J. (2001). "Searching for the switch: neural bases for perceptual rivalry alternations." Brain and Mind 2: 85-118.
- Tong, F., Meng, M. & Blake, R. (2006) Neural bases of binocular rivalry. Trends in Cognitive Sciences. 10, 502-511. 
- Wade, N. (1998). A natural history of vision. MIT Press, Cambridge MA.
- Walker, P. (1978) Binocular rivalry: Central or peripheral selective process? Psychological Bulletin, 85, 376-389.
- Naotsugu Tsuchiya and Christof Koch (2008) Attention and consciousness. Scholarpedia, 3(5):4173.
- Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.
- Olaf Sporns (2007) Complexity. Scholarpedia, 2(10):1623.
- Keith Rayner and Monica Castelhano (2007) Eye movements. Scholarpedia, 2(10):3649.
- Naotsugu Tsuchiya (2008) Flash suppression. Scholarpedia, 3(2):5640.
- Giovanni Gallavotti (2008) Fluctuations. Scholarpedia, 3(6):5893.
- William D. Penny and Karl J. Friston (2007) Functional imaging. Scholarpedia, 2(5):1478.
- Seiji Ogawa and Yul-Wan Sung (2007) Functional magnetic resonance imaging. Scholarpedia, 2(10):3105.
- Florian Mormann and Christof Koch (2007) Neural correlates of consciousness. Scholarpedia, 2(12):1740.
- Peter Jonas and Gyorgy Buzsaki (2007) Neural inhibition. Scholarpedia, 2(9):3286.
- Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.
- Jeff Moehlis, Kresimir Josic, Eric T. Shea-Brown (2006) Periodic orbit. Scholarpedia, 1(7):1358.
- John Dowling (2007) Retina. Scholarpedia, 2(12):3487.
- Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.
- Arkady Pikovsky and Michael Rosenblum (2007) Synchronization. Scholarpedia, 2(12):1459.
- Anthony T. Barker and Ian Freeston (2007) Transcranial magnetic stimulation. Scholarpedia, 2(10):2936.
- Dale Purves, William T. Wojtach, Catherine Howe (2008) Visual illusions: An Empirical Explanation. Scholarpedia, 3(6):3706.
- Bruno G. Breitmeyer and Haluk Ogmen (2007) Visual masking. Scholarpedia, 2(7):3330.
- Alais, D. & Blake, R. (2005) Binocular rivalry and perceptual ambiguity, MIT Press, Cambridge MA.
- Kim, C.Y. & Blake, R. (2005) Psychophysical magic: rendering the visible 'invisible'. Trends in Cognitive Science 9: 381-388
- Maier, A. et al. (2008) "Divergence of fMRI and neural signals in V1 during perceptual suppression in awake monkey." Nature Neuroscience 11: 1193-1200