Encyclopedia of Color Science and Technology

2016 Edition
| Editors: Ming Ronnier Luo

Binocular Color Matching

Reference work entry
DOI: https://doi.org/10.1007/978-1-4419-8071-7_81

Synonyms

Definition

In the context of this article, “binocular color matching” refers to color vision with two eyes, focusing on the differences between when two eyes are used to view colors (i.e., binocular or dichoptic color) and when only one is used (i.e., monocular color), as well as phenomena (largely) unique to binocular vision such as fusion, rivalry, and stereopsis.

Introduction

Most of us have two functioning eyes, and yet we still somehow see the visual world as a single entity, most of the time unaware that our brains are receiving and processing two separate and largely independent sets of visual signals from the outside world. In the context of color vision, this raises some interesting questions:
  1. 1.

    Binocular color fusion: How similar do the chromatic properties of visual signals in each eye have to be in order to be perceived as one?

     
  2. 2.

    Binocular color rivalry: How different do the chromatic properties of visual signals in each eye have to be in order to evoke binocular rivalry, which is an alternation in visibility of the visual information going to each eye?

     
  3. 3.

    Color and stereopsis: Stereopsis is the perception of depth that is obtained by analyzing the differences between the images going to each eye. To what extent is this perception of depth affected by the chromatic properties of visual information?

     
  4. 4.

    Binocular color summation: The processing of visual signals is affected by whether or not a scene is viewed with one or two eyes: to what extent do the chromatic properties of visual signals influence this?

     
  5. 5.

    Binocular color appearance: Do colors look different when viewed with one eye or two?

     

Note that the field of binocular vision and stereopsis has been thoroughly reviewed by Prof. Brian Rogers of the University of Oxford, UK, and Prof. Ian Howard, late of York University, Canada [11], and for all but the most recent studies, the interested reader should refer to them for further information.

Binocular Color Fusion

The problem of binocular fusion, how the world is seen as one despite having two eyes, has fascinated philosophers and scientists for centuries (see [11] for a historical overview). It makes sense that color might have a role to play in this phenomenon. One might expect a feature in the visual world when viewed with one eye to have the same spectral reflectance properties and therefore presumably appear to be the same color, when viewed with the other eye. Consequently, interocular color similarity would presumably make a good visual cue for fusion. Curiously, however, most of the research effort has gone into investigating what happens when the colors presented to the two eyes are different. In particular, there is the controversial issue of dichoptic color mixture: when two lights with different chromatic contents are presented separately to each eye, do the lights combine in a similar way to when they are presented superimposed to the same eye or does binocular rivalry (an alternation in perception between the stimuli presented to each eye) result (see Fig. 1)? The first investigation of this phenomenon appears to have been in the early eighteenth century using differently colored silks viewed through an aperture [7]. Later, in the nineteenth century, the existence or not of dichoptic color mixture was a further source of controversy between Helmholtz and Hering, given the different predictions of the Young-Helmholtz and Hering theories of color vision. Arguably the most controversial aspect is whether or not “binocular yellow” can be obtained when red light is presented to one eye and green to the other. Recent studies have clarified that true dichoptic color mixture does appear to take place, so long as certain conditions are met. These are that the mixture is more stable with small and textured patches of light rather than large and homogeneous, with flickering rather than steady stimuli, and with patches of low luminance and saturation which are close in luminance and chromaticity rather than high luminance and saturation with different luminances and chromaticity ([11]; see also [10]). Some of these rules are nicely illustrated in a study by de Weert and Wade [5] (see Fig. 2). In a more recent study, the bold claim is made that “two eyes are worse than one,” based on data suggesting that monocularly visible features of different colors effectively disappear when dichoptically combined [1]. In an earlier study by Simmons [32], however, although detection thresholds for briefly presented isoluminant red-green gratings in antiphase between the eyes were higher than those for monocular presentation, they were not so high that the signals from each eye were effectively canceling each other out, as suggested by Anstis and Rogers [1]. Potentially, technical issues to do with the accurate registration of stimuli in each eye and the presence of barely detectable luminance artifacts complicate the interpretation of these experiments.
Binocular Color Matching, Fig. 1

Schematic diagram illustrating alternative percepts generated by dichoptic color mixing or binocular rivalry

Binocular Color Matching, Fig. 2

Dichoptic color mixing: fusion of the upper two disks gives unstable binocular rivalry; fusion of the lower two textured disks produces stable dichoptic color mixing. This illustrates the role of texture in promoting dichoptic color mixing (Taken from de Weert and Wade [5], Reprinted from Vision Research, 28, Charles M.M. de Weert, Nicholas J. Wade, Compound binocular rivalry, 1031–1040, Copyright (1988), with permission from Elsevier)

A different approach was taken by Malkoc and Kingdom [21], who, following on from an earlier study by Yoonessi and Kingdom [42], measured dichoptic color difference thresholds (DCDTs), which are the thresholds for detecting a color difference between the two eyes, coinciding with the detection of a peculiar phenomenon called binocular luster. They found that these thresholds were higher than those for detecting color differences between two stimuli when they were presented side by side, although lower than those required to provoke binocular rivalry (see below). Malkoc and Kingdom [21] also found that these thresholds were best predicted by the perceived color difference between the two stimuli, rather than any considerations based on cardinal or unique hue mechanisms. Kingdom and Libenson [14] specifically investigated the processing of interocular differences in saturation (or color contrast). They found that the appearance of the mixture obtained crucially depended on the relative amounts of luminance and chromatic contrast. With purely chromatic differences between the two eyes (i.e., lights with the same hue but different saturations), the more saturated/higher contrast stimulus dominated the percept, but the presence of a luminance pedestal forced the colors to blend and therefore reduced the saturation of the resultant. Kingdom and Libenson [14] argue from these results that the appearance of a dichoptic color mixture depends on whether or not the brain interprets the information from the two eyes as coming from the same object or not, a phenomenon which they term the “object commonality hypothesis.”

Binocular Color Rivalry

As rivalry is, to a certain extent, the obverse of fusion, most of the issues pertinent to this theme have already been discussed above. It is certainly true that the presentation of saturated red stimuli to one eye and saturated green to the other is often almost paradigmatic in studies where the aim is to evoke binocular rivalry (see, e.g., [2]; also, again, [5]). O’Shea and Williams [23] demonstrated that S-cone-isolating stimuli could induce binocular rivalry, suggesting that rivalry was not solely confined to luminance or red-green chromatic pathways. A detailed study of the wavelength sensitivity of binocular rivalry was performed by Sagawa [28]. Note that rivalry does not occur for briefly presented stimuli. In these situations the dichoptic stimuli tend to superimpose, although are still distinguishable from monocularly superimposed stimuli [11]. There is some evidence that the chromatic system is more affected by binocular rivalry suppression than the achromatic system [11, 24]. Mullen et al. [22] found that the visibility of chromatic grating stimuli presented to one eye was affected by the presence of luminance stimuli in the other, suggesting that, when stimuli differ between the two eyes, the suppression of one eye by the other is independent of whether the stimulus contains color or luminance contrast (which is not the case for monocular vision, or when the stimulus is the same in both eyes, when the suppression of one stimulus by another, masking, stimulus is more selective).

Color and Stereopsis

There are two issues which have dominated research on color and stereopsis. The first is whether or not there is actually a functioning chromatic stereopsis mechanism. The second is whether or not chromatic information helps to solve the stereo correspondence problem. In both cases the arguments have been similar to those in color motion perception. Historically, a good way of testing whether or not a purely chromatic mechanism exists has been to test performance at isoluminance, when the visual patterns in question are theoretically defined solely by chromatic contrast, without any luminance contrast being present. The earliest study of stereopsis at isoluminance suggested that stereoscopic depth perception was very weak and possibly nonexistent at red-green isoluminance [20]. Subsequent studies provided conflicting results, depending on the precise stimulus characteristics [11]. Around the early 1990s, the extreme views on this issue were represented by two studies: Livingstone and Hubel [19] and Scharff and Geisler [29]. Livingstone and Hubel [19] claimed that all previous demonstrations of intact stereoscopic depth perception were artifactual, and down to the technical difficulties associated with removing all luminance information from the stimulus. Their view that stereopsis was essentially “color blind” was consistent with their theories on the parallel processing of visual information by magnocellular- and parvocellular-mediated visual pathways, with stereopsis mediated by the “color-blind” magnocellular stream. Scharff and Geisler [29], on the other hand, claimed that their careful calibration of the stimulus, accounting for all potential luminance artifacts, not only demonstrated that stereopsis was possible at isoluminance but that it was as good as it could be, given the limits on obtainable color contrast due to the overlap in spectral sensitivities of long- and medium-wavelength-sensitive cones. In a series of studies published between 1994 and 2002, Simmons and Kingdom updated this view of the status of stereopsis at isoluminance to demonstrate that there exists a rudimentary chromatic stereopsis mechanism which is less contrast sensitive and has a more limited disparity range, poorer stereoacuity, and poorer ability to encode a stereoscopically defined shape than its luminance counterpart ([15, 17, 33, 34, 35]; reviewed in [16]). While this view was challenged in a study by Krauskopf and Forte [43], who argued, similarly to Livingstone and Hubel [19], for a complete disappearance of stereopsis at isoluminance, their data could be partially explained by the presence of high spatial frequency luminance artifacts in their stimuli. Having said that, it is very difficult to control for these artifacts, especially given the high contrast sensitivity of luminance-based stereopsis [31, 37]. While most of the above studies were carried out at red-green isoluminance, Grinberg and Williams [9] demonstrated that stereopsis is also possible at “blue-yellow” isoluminance.

Can chromatic information help to solve the stereo correspondence problem? The correspondence problem in stereopsis refers to the ability of the human brain to work out which feature in one eye’s view matches with which feature in the other eye’s view (see Fig. 3). As stated above, it makes sense that chromatic information could help in solving this problem, and, sure enough, a number of studies have demonstrated that this is indeed the case (e.g., [6, 12, 18]). Simmons and Kingdom [36] argued that these results might be explicable in terms of the interactions between different independent stereopsis mechanisms, some of which are sensitive to chromatic contrast (see also [38]). Wardle and Gillam [41] have argued that color information is also important in the so-called da Vinci stereopsis, which is the perception of depth obtained, under certain viewing conditions, from visual regions only visible to one eye.
Binocular Color Matching, Fig. 3

Schematic illustration of the potential role of color in the stereo correspondence problem. If the images illustrated are presented to the left and right eyes, respectively, if color promotes stereo correspondence, then a green bar will appear in front and a red bar behind the fixation plane (upper right). If color has no role, then four bars will be seen in the frontoparallel plane (lower right). The bars colored yellow might also appear rivalrous, depending on the precise stimulus conditions

There are some complicated dependences of stereoscopic sensitivity on stimulus dynamics at isoluminance which were investigated by Tyler and Cavanagh [39]. Note that the phenomenon of chromostereopsis, most often experienced as a saturated red target appearing in front of a saturated blue one (when viewed with two eyes) despite the stimuli being physically in the same depth plane, is thought to be largely due to chromatic aberrations in the eye inducing a relative difference (i.e., disparity) in the retinal locations of the red and blue objects in each eye [11, 25]. Figure 4, especially if presented on a data projector and viewed in a dark room, normally induces a strong sensation of this illusory depth percept.
Binocular Color Matching, Fig. 4

Stimulus for chromostereopsis. If viewed on the full screen (especially if presented on a data projector), most people will see the red “CHROMO” look slightly in front of the blue “STEREOPSIS”

Binocular Color Summation

To what extent does viewing a chromatic stimulus with two eyes affect its visibility or the performance obtained with it? The scientific literature presents a somewhat confusing picture, with Anstis and Rogers [1] claiming that “two eyes are worse than one,” Tyler and Cavanagh [39] claiming that “two eyes” are “as sensitive as one,” and Simmons [32] finding evidence for binocular contrast summation close to “full summation,” when contrast is effectively linearly added between the eyes, and so a binocular stimulus is twice as detectable as when presented to both eyes rather than one (this would be “two eyes are twice as good as one”!). The conditions under which dichoptic color mixing is obtained have already been outlined above. It therefore suffices to say that the extent of binocular summation for a given chromatic stimulus will depend on the task and the precise chromatic and spatiotemporal parameters of the stimulus.

Binocular Color Appearance

Those of us who have typical binocular vision will have a region of binocular overlap in the middle of the visual field flanked on either side by regions that are only visible in one or the other eye. However, we are usually not aware that colors change in the center versus the periphery of our visual field, although changes are detectable under controlled laboratory conditions [3, 26]. A whole range of dichoptic mechanisms have been proposed to account for these apparent compensations between the two eyes. Shimono et al. [30] suggested a model which takes into account previous results on interocular transfer of chromatic adaptation and chromatic induction effects (e.g., [4, 8, 40]) by proposing a “purely binocular” chromatic mechanism, which responds only to simultaneous binocular presentation of similar colors to each eye but which also inhibits the responses of “purely monocular” color mechanisms (see Fig. 5). This model would also seem to be reasonable in the light of physiological data collected from monkey visual cortex [27]. However, more recent results, such as those of Kingdom and Libenson [14] and Mullen et al. [22] (discussed above), complicate this simple picture.
Binocular Color Matching, Fig. 5

Model from Shimono et al. [30] which illustrates the inhibitory action of their proposed “purely chromatic” binocular mechanism on monocular mechanisms (Taken from Shimono et al. [30], Reprinted from Vision Research, 49, Shimono, K., Shioiri, S., Yaguchi, H., Psychophysical evidence for a purely binocular color system, 202–210, Copyright (2009), with permission from Elsevier)

Another aspect of binocular color appearance that has recently received attention is how presenting different colors to the two eyes can be an easy way to simulate surface glossiness in stereoscopic visual displays, without the need to create a detailed reflectance model [13].

Conclusion

The field of binocular color vision has proven surprisingly controversial over the years, with the existence of true dichoptic color mixing, the contribution of color to stereopsis, and the presence or not of a purely binocular color mechanism being key battlegrounds. The current consensus of the field appears to be that dichoptic color mixing can happen, that chromatic mechanisms contribute to stereopsis, and that there is a purely binocular color mechanism, but demonstrating these things requires careful stimulus control and experimental technique. Added to this there are some curious phenomena such as being able to make stimuli which are easily detectable when viewed monocularly disappear when viewed binocularly [1] and the interocular transfer of chromatic induction [8] (Fig. 6). It seems that we still have some more work to do before we can say that we fully understand the complex interactions between binocular color mechanisms.
Binocular Color Matching, Fig. 6

Binocular color induction in monocular patches. Sets of multicolored stereograms are shown for three different color combinations. (a) The white, monocular patches gradually fill with distinct pastel colors during binocular viewing of the stereograms. The buildup of the effect typically takes a few seconds. Dependent on the viewer, the full development of the colors may even take minutes. The color effects are most compelling when the stereograms are viewed printed out and in bright light (see original publication). When one eye is closed, the illusion gradually becomes weaker and disappears. This buildup and decay of the illusion after opening or closing one eye indicates that the color appearances are induced by a binocular mechanism. (b) During binocular viewing, the white, binocular patches do not change color. During monocular viewing after a period of binocular viewing, however, induced colors are apparent in the patches (Taken from Erkelens and Van Ee [8], Reprinted from Vision Research, 42, Erkelens, C.J., Van Ee, R., Multi-coloured stereograms unveil two binocular colour mechanisms in human vision, 1103–1112, Copyright (2002), with permission from Elsevier)

Cross-References

References

  1. 1.
    Anstis, S., Rogers, B.: Binocular fusion of luminance, color, motion and flicker – two eyes are worse than one. Vision Res. 53, 47–53 (2012)CrossRefGoogle Scholar
  2. 2.
    Bhardwaj, R., O’Shea, R.P.: Temporal analysis of image-rivalry suppression. PLoS One 7(9), e45407 (2012). doi:10.1371/journal.pone.0045407ADSCrossRefGoogle Scholar
  3. 3.
    Bompas, A., Powell, G., Sumner, P.: Systematic biases in adult color perception persist despite lifelong information sufficient to calibrate them. J. Vis. 13, 1–19 (2013)CrossRefGoogle Scholar
  4. 4.
    Coltheart, M.: Color-specificity and monocularity in the visual cortex. Vision Res. 3, 2595–2598 (1973)CrossRefGoogle Scholar
  5. 5.
    De Weert, C.M.M., Wade, N.J.: Compound binocular rivalry. Vision Res. 28, 1031–1040 (1988)CrossRefGoogle Scholar
  6. 6.
    Den Ouden, H.E.M., Van Ee, R.R., de Haan, E.H.E.: Colour helps to solve the binocular matching problem. J. Physiol. 567, 665–671 (2005)CrossRefGoogle Scholar
  7. 7.
    Desagulier, J.T.: A plain and easy experiment to confirm Sir Isaac Newton’s doctrine of the different refrangibility of the rays of light. Philos. Trans. R. Soc. Lond. A 29, 448–452 (1716)CrossRefGoogle Scholar
  8. 8.
    Erkelens, C.J., Van Ee, R.: Multi-coloured stereograms unveil two binocular colour mechanisms in human vision. Vision Res. 42, 1103–1112 (2002)CrossRefGoogle Scholar
  9. 9.
    Grinberg, D.L., Williams, D.R.: Stereopsis with chromatic signals from the blue-sensitive mechanism. Vision Res. 25, 531–537 (1985)CrossRefGoogle Scholar
  10. 10.
    Hovis, J.K.: Review of dichoptic color mixing. Optom. Vis. Sci. 66, 181–190 (1989)CrossRefGoogle Scholar
  11. 11.
    Howard, I.P., Rogers, B.J.: Perceiving in Depth: volume 2 Stereoscopic Vision. Oxford University Press, Oxford, UK (2012)CrossRefGoogle Scholar
  12. 12.
    Jordan, J.R., Geisler, W.S., Bovik, A.C.: Color as a source of information in the stereo correspondence process. Vision Res. 30, 1955–1970 (1990)CrossRefGoogle Scholar
  13. 13.
    Jung, W.-S., Moon, Y.-G., Park, J.-H., Song, J.-K.: Glossiness representation using binocular color difference. Opt. Lett. 38, 2584–2587 (2013)ADSCrossRefGoogle Scholar
  14. 14.
    Kingdom, F.A.A., Libenson, L.: Dichoptic color saturation mixture: binocular luminance contrast promotes perceptual averaging. J. Vis. 15(5), Article no. 2. (2015)Google Scholar
  15. 15.
    Kingdom, F.A.A., Simmons, D.R.: Stereoacuity and colour contrast. Vision Res. 36, 1311–1319 (1996)CrossRefGoogle Scholar
  16. 16.
    Kingdom, F.A.A., Simmons, D.R.: The relationship between colour vision and stereoscopic depth perception (invited paper). J. Soc. 3D Broadcast. Imaging 1, 10–19 (2000)Google Scholar
  17. 17.
    Kingdom, F.A.A., Simmons, D.R., Rainville, S.R.: On the apparent collapse of stereopsis in random-dot stereograms at isoluminance. Vision Res. 39, 2127–2141 (1999)CrossRefGoogle Scholar
  18. 18.
    Kovacs, I., Julesz, B.: Depth, motion, and static-flow perception at metaisoluminant color contrast. Proc. Natl. Acad. Sci. U. S. A. 89, 10390–10394 (1992)ADSCrossRefGoogle Scholar
  19. 19.
    Livingstone, M.S., Hubel, D.H.: Psychophysical evidence for separate channels for the perception of form, colour, movement, and depth. J. Neurosci. 7, 3416–3468 (1987)Google Scholar
  20. 20.
    Lu, C., Fender, D.H.: The interaction of color and luminance in stereoscopic vision. Invest. Ophthalmol. 11, 482–489 (1972)Google Scholar
  21. 21.
    Malkoc, G., Kingdom, F.A.A.: Dichoptic difference thresholds for chromatic stimuli. Vision Res. 62, 75–83 (2012)CrossRefGoogle Scholar
  22. 22.
    Mullen, K., Kim, Y.J., Gheiratmond, M.: Contrast normalization in colour vision: the effect of luminance contrast on colour contrast detection. Sci. Rep. 4, 7350 (2014). doi:10.1038/srep07350ADSCrossRefGoogle Scholar
  23. 23.
    O’Shea, R.P., Williams, D.R.: Binocular rivalry with isoluminant stimuli visible only via short-wavelength-sensitive cones. Vision Res. 36, 1561–1571 (1996)CrossRefGoogle Scholar
  24. 24.
    Ooi, T.L., Loop, M.S.: Visual suppression and its effect upon color and luminance sensitivity. Vision Res. 34, 2997–3003 (1994)CrossRefGoogle Scholar
  25. 25.
    Ozolinsh, M., Muizniece, K.: Color difference threshold of chromostereopsis induced by flat display emission. Front. Psychol. 6, Article: 337 (2015)Google Scholar
  26. 26.
    Parry, N.R.A., Panorgias, A., McKeefry, D.J., Murray, I.J.: Real-world stimuli show perceived hue shifts in the peripheral visual field. J. Opt. Soc. Am. A 29, 96–101 (2012)ADSCrossRefGoogle Scholar
  27. 27.
    Peirce, J.W., Solomon, S.G., Forte, J.D., Lennie, P.: Cortical representation of color is binocular. J. Vis. 8, 1–10 (2008)CrossRefGoogle Scholar
  28. 28.
    Sagawa, K.: Minimum light intensity required for color rivalry. Vision Res. 21, 1467–1474 (1981)CrossRefGoogle Scholar
  29. 29.
    Scharff, L.V., Geisler, W.S.: Stereopsis at isoluminance in the absence of chromatic aberrations. J. Opt. Soc. Am. A 9, 868–876 (1992)ADSCrossRefGoogle Scholar
  30. 30.
    Shimono, K., Shioiri, S., Yaguchi, H.: Psychophysical evidence for a purely binocular color system. Vision Res. 49, 202–210 (2009)CrossRefGoogle Scholar
  31. 31.
    Simmons, D.R.: The minimum contrast requirements for stereopsis. Perception 27, 1333–1343 (1998)CrossRefGoogle Scholar
  32. 32.
    Simmons, D.R.: The binocular combination of chromatic contrast. Perception 34, 1035–1042 (2005)CrossRefGoogle Scholar
  33. 33.
    Simmons, D.R., Kingdom, F.A.A.: Contrast thresholds for stereoscopic depth identification with isoluminant and isochromatic stimuli. Vision Res. 34, 2971–2982 (1994)CrossRefGoogle Scholar
  34. 34.
    Simmons, D.R., Kingdom, F.A.A.: Differences between stereopsis with isoluminant and isochromatic stimuli. J. Opt. Soc. Am. A 12, 2094–2104 (1995)ADSCrossRefGoogle Scholar
  35. 35.
    Simmons, D.R., Kingdom, F.A.A.: On the independence of chromatic and achromatic stereopsis mechanisms. Vision Res. 37, 1271–1280 (1997)CrossRefGoogle Scholar
  36. 36.
    Simmons, D.R., Kingdom, F.A.A.: Interactions between chromatic- and luminance-contrast-sensitive stereopsis mechanisms. Vision Res. 42, 1525–1546 (2002)CrossRefGoogle Scholar
  37. 37.
    Simmons, D.R.: Stereopsis at red-green isoluminance: chasing the luminance artifacts. [Abstract]. J. Vis. 2(10), 69a (2002). doi:10.1167/2.10.69. http://journalofvision.org/2/10/69/
  38. 38.
    Stuart, G.W., Edwards, M., Cook, M.L.: Color inputs to random-dot stereopsis. Perception 21, 717–729 (1992)CrossRefGoogle Scholar
  39. 39.
    Tyler, C.W., Cavanagh, P.: Purely chromatic perception of motion in depth: two eyes as sensitive as one. Percept. Psychophys. 49, 53–61 (1991)CrossRefGoogle Scholar
  40. 40.
    Vimal, R.L.P., Shevell, S.K.: A central binocular mechanism affects chromatic adaptation. Vision Res. 27, 429–439 (1987)CrossRefGoogle Scholar
  41. 41.
    Wardle, S.G., Gillam, B.J.: Color constrains depth in da Vinci stereopsis for camouflage but not occlusion. J. Exp. Psychol. Hum. Percept. Perform. 39, 1525–1540 (2013)CrossRefGoogle Scholar
  42. 42.
    Yoonessi, A., Kingdom, F.A.A.: Dichoptic difference thresholds for uniform color changes applied to natural scenes. J. Vis. 9(2), 1–12 (2009) (article no. 3)CrossRefGoogle Scholar
  43. 43.
    Krauskopf, J., Forte, J.D.: Influence of chromaticity on vernier and stereo acuity. J. Vis. 2(9), 645–652 (2002)Google Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.School of PsychologyUniversity of GlasgowGlasgowUK