Information

If longer wavelength color is perceived faster by human, what about non-spectral color (black)?

If longer wavelength color is perceived faster by human, what about non-spectral color (black)?


We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

Related to human vision, I read the hypothesis about how longer-wavelength color is perceived faster by human eye than shorter-wavelength color source : A Brief Classification of Colour Illusions. But what if the other color is a non-spectral color like, black ? Black is achromatic color, it's the absence of colors and has no wavelength. Does this perception hypothesis still apply ?

Because I've been doing a color comparison by putting a layer of colors on top of each other. In this case I tried with black - green (~549nm) and black - yellow (~570nm), and it seems like black is always dominant against other colors. But I can't find any scientific reference for this. The closest I found is the hypothesis I mentioned before. Any suggestion ?


Ingenious: Mazviita Chirimuuta

W hen I asked Mazviita Chirimuuta why philosophers were so crazy about color, she smiled, as if she sensed my question had an implied criticism of philosophy’s penchant for chewing more than it bit off, to paraphrase what a wit once said of Henry James. But then she offered an eloquent answer. “Philosophers are motivated to study color in detail because the senses are our means to understanding the world,” she said. “Does that mean all of our knowledge is subjective? And if so, is that a bad thing? Underlying the debate about color is this bigger question about the nature of knowledge, which comes to us by means of our senses.”

Philosophers and scientists have been parsing the meaning of color since at least the heady 1600s when Newton “stuck a needle-like bodkin in his eye and pressed with it until he saw white and colored circles,” as author Leonard Mlodinow kindly informed us. With somewhat less pain, physicists demonstrated color is determined by wavelengths and surface reflection. Physiologists detailed how specific anatomy in our eyes and brains tells us what is blue or red. In her recent book, Outside Color, her first, Chirimuuta boldly concludes that color should be defined not by the world outside or inside our heads, but an interaction of both.

Chirimuuta, 37, is an assistant professor in the Department of History and Philosophy of Science at the University of Pittsburgh. She has been captivated by the subjectivity of color since she was a kid, when, she explains in our interview, she had one of those existential kid questions about the world: What is the real color of the ground? Growing up in England, her mother, an ophthalmologist, fostered her curiosity in science. (Her father is from Zimbabwe.) In high school, she read The Man Who Mistook His Wife for a Hat, by Oliver Sacks, and was struck by “how people’s whole worlds could be changed and warped in completely unpredictable ways.” She grew up to earn a Ph.D. in visual neuroscience from the University of Cambridge.

Our interview unreeled inside the University of Pittsburgh’s gothic Cathedral of Learning. As an academic in philosophy and neuroscience, Chirimuuta said, she wasn’t accustomed to speaking to a popular science audience. Did she have to worry about using philosophical or technical language? I assured her that being herself was the best policy, and as the interview reveals, she didn’t have to worry at all. Her explanations were thoughtful and articulate, free of the clouds of jargon.

The video plays at the top of the screen.

Flashback: Human Uniqueness

A physicist and a philosopher walk into a lab… no, this isn’t the start of a joke. It’s an everyday occurrence in the lab of Andrew Briggs, Professor of Nanomaterials at Oxford University. While working on how to exploit quantum. READ MORE

How does a physicist typically explain color?

A lot of people have the idea that if you look at color from the perspective of physics, then color is simply wavelength of light. Everyone has seen Newton’s spectrum—if you take white daylight and split it into all of its different component colors, you get a rainbow, and each color in that rainbow has a specific wavelength of light. So people think, “Oh well, color is just wavelength of light and in the world around us, different objects reflect different wavelengths of light therefore we see them as having different colors.” But actually, it’s not like that.

Because if you actually look at the wavelengths that are reflected off say, red objects, or blue objects, they’re not ever one specific wavelength that we always experience and say “oh, that’s blue, that’s just that one wavelength of light.” In fact, depending on the context, the same wavelength of light could be seen as different colors because of color constancy, so the brain, or our visual system, is always taking into account what’s going on in the surrounding environment. The color of the light that is illuminating objects will affect the color that we perceive the object to have in some cases and in some cases not so there’s no one-to-one correlation between wavelength of light and the color we perceive.

How does a neurobiologist explain color?

Someone who has ordinary human color vision has three kinds of different cone receptors in their retina. The cones are the photoreceptors that are sensitive during daylight-level illumination, so when the world is bright enough for us to see colors well, our cones are operating and they have sensitivities to wavelength which differ from each other, so they provide the very first input to the color visual system.

Because our cones respond differently to wavelengths, we see objects as having different colors, but again, there’s no one-to-one relationship—“your long wavelength receptor is operating, therefore you see red”—it really depends on the balance between what the different cones are receiving from the light in the world around them and then how the brain interprets that in terms of the context of the whole scene.

What are the “outer” and “inner” views of color?

There’s a long tradition in philosophy that sees the “outer” and “inner” as an absolute dichotomy and I trace that back to, at least, the 17th century when science was being developed in the way that we now know it—the idea of physics being the science of things as they are, independent of the human mind. Then the question arises: There are some parts of experience of the world which seem to be subjective, or at least informed by or shaped by the human mind, so what do we say about those? Are they purely subjective? Is it just some part of our inner consciousness and therefore, do we make this radical break between physics and psychology?

So that kind of dualistic approach, that’s always associated with René Descartes. He famously proposed mind-brain dualism and said that the mind was a completely different substance from the material world, but I think that tendency to see the inner world as something separate from the outer world is more pervasive than just a theory of mind-brain dualism—it’s really a way of trying to understand nature as something entirely objective. Then there’s this problem about how we fit the mind back in, and I think that’s something that will take a lot of philosophical work to really think through in a more satisfactory way.

What is your theory of color?

Going back to that problem about this dichotomy between the inner and outer, there’s been this tendency to say, “well, anything that’s subjective in our knowledge or in our experience isn’t on the same footing as things that we know about as being completely objective.” I say that this sort of difficulty that people have of accepting subjective aspects of experience and knowledge then leaves people to say well, “if color’s not part of physics, then it must be a complete illusion” and I’m saying well, actually we need a way of theorizing subjectivity in such a way that we’ll just acknowledge that there are parts of our experience and our perceptual knowledge of things that are generated by the particular ways that we interact with the world.

As humans who have three kinds of photoreceptors, or two, or sometimes one kind of photoreceptor for daylight vision—that means we interact with the world in a particular way that informs our experience of the world. If our visual systems were built differently, our whole visual experience, and probably our knowledge of the world, would be quite different but there’s nothing inherently problematic about that. So I think of color as a property, or something that can only be understood in terms of the particular ways that we interact with the world. That’s my way of saying that we should try and see those inner and outer domains as not as separate from each other as we think. Really, there’s this constant back and forth between the two and that’s how visual experience is generated, and knowledge maybe more generally.

How have you changed the debate about color?

The way the debate has standardly gone is to say, “well, if color is anything, if color exists, then it’s a property of objects.” So if you’re a realist, you’ll say “yeah, the maroon property belongs to this seat the whiteness property belongs to that wall.” If you’re an anti-realist, you’ll just say “no, no objects have those properties color doesn’t exist.” What I’ve done is say that actually, a better way of thinking about color is not as a property of objects, but as a property of interactions that perceivers have with objects. In my view, which is the view I call color adverbialism, there are perceptual processes that are going on all the time. Every time we look around a room, light’s bouncing off the walls into my eyes and my brain’s processing this information and I’m saying that that whole extended interaction between myself and my surroundings, that’s the thing that has color, not the objects that I see. So when I talk about what’s there in my surroundings, I say that color is my way of seeing those things, so I see that wall in a white way, so really the whiteness is modifying that perceptual experience. It’s more a property of the experience or that process, that activity I’m doing, rather than the wall itself.

What’s the evolutionary purpose of color?

That’s a really fascinating debate that’s ongoing within the science. One idea about the particular color visual systems that primates like us have—the trichromatic visual system that we have—people have speculated that well, having the different long and medium wavelength receptors in our retina, that helps us distinguish ripe fruit from green foliage. So there’s this idea that different animals have particular kinds of visual systems that make ecologically relevant objects more apparent to them. If an animal needs to feed from a specific kind of tree, evolution will engineer a visual system that will help them see those things. Bees have very good color vision they use it for foraging, for finding pollen from flowers.

There are all these explanations and stories one could tell. It’s very hard to ever find smoking guns with these things, but there’s good reason to think that the reason why different animals have very different kinds of color visual systems is related to the particular kinds of tasks that they’re engaged with, whether it’s feeding or navigating or escaping from predators.

Do big-brain humans have a richer experience of color than other animals?

No, because an interesting thing is that most animals that aren’t mammals have more interesting and more exotic color visual systems than we have, so birds typically have many more kinds of photoreceptors, like five there are some kinds of sea creatures, like the mantis shrimp, that have kinds of color receptors that go into the teens. If we could compare our color visual experience with that of a pigeon or a mantis shrimp, it would probably seem quite boring.

What do you mean by “perception of color”?

Typically when we talk about perception of color, we talk about the conscious visual experiences that we have, because if you have vision, then color is such a dramatic part of our conscious experience of the world. Most of the color debate is really framed in terms of these very bright, catchy, qualitative experiences that we have of red and yellow and all the colors of the rainbow.

But that said, most of the processing behind color visual experience is unconscious, like when we have color constancy and there are lots of computational theories about how color constancy works. We can’t introspect on that, that’s all unconscious to us. We just know its results.

Why did people see the notorious dress as different colors?

There’s a kind of ready-at-hand explanation that comes from thinking about color constancy. If you take that image of the dress, it depends on whether you think that the dress has been lit by a [shadowy] bluish colored light—if you interpret the scene that way, then you’ll see it as white and gold if you think it’s not in that kind of lighting, you’ll see it as blue and black. So people have explained it in terms of what assumption each individual is making about the lighting in the wider scene, which is why the photograph itself is ambiguous, because the lighting in the scene can be interpreted in different ways.

This is what I mean about color constancy being completely unconscious to us, but that’s how vision scientists model it. It’s like there’s a little algorithm in your brain checking what the illumination is doing and the direction of lighting and what color it is, and then balancing that against the object that’s seen and it just seems that different people’s inner algorithms work slightly differently with the dress, which is why you get this disagreement.

Has our experience of color changed over time?

In terms of talking about experience and how it’s become different over time, I think it’s really interesting to look at the cultural evolution of color. If we just look at the history of Western civilization, or if we look at texts from ancient Greece, how color is written about is very different from how we would write and think about color now, which suggests that people experienced the visual world—at least to describe it—in a way that was quite different. For example, they would group colors, certain colors, together that we would think are different. A famous phrase in Homer, “the wine dark sea”—when we read that we think well, the sea’s blue, it’s not red. But the word that he was using was probably for some dark shade that wasn’t associated with blue or red and those hues, but actually was more about the tonality of color whereas, our color concepts are much more tied to hues, you know, those color boxes and those crayons that we remember from children. That’s our primary means of grasping color concepts—it’s specific rainbow shades whereas other cultures pay more attention to light and darkness and also material qualities of color like how shiny and dull they are.

What inspired you to study color?

Okay, so it will sound like I’m making this up! But when I was a child, I remember looking at the ground—and this was a concrete playground area around me, so it was this gray concreted surface and some of it was in shade and some of it wasn’t—and I remember asking myself, what’s the real color of the ground? Which is basically, this problem of perceptual relativity and the subjectivity of color that is debated up until now. It’s not like I went through my whole education thinking to myself I will study this one day, but it was one of the reasons why I became interested in vision.

My mother is an ophthalmologist and I remember also, around the same age, asking her about how seeing works, and looking at children’s encyclopedias where they show you images projecting to the retina and being upside down and thinking, well how come the world doesn’t look upside down? And these are just the kinds of things that get people who eventually turn out to be a philosopher sort of fascinated by perception.

What sparked your interest in science?

When I was in high school, my physical science teachers were really inspiring. I had an amazing chemistry teacher who went through the periodic table, explaining how the different elements bound together, and different properties and patterns in nature, and then I also had a great physics teacher. Learning about how to use equations to explain phenomena in the natural world and doing experiments. I still look back at those lessons—they were very inspiring. I’m a big fan of great early science education in schools because I think it’s an amazing gift to give a child: some of that firsthand knowledge of nature and then to enthuse them to carry on studying more.

What would you be if you weren’t a scientist?

I think I would always want to write, so maybe a different kind of writer.


5 Answers 5

Color is a double valued concept, different in physics and in perception.

In Physics there is a one to one correspondence between color seen in the visible spectrum and the frequency of light.

The whole electromagnetic spectrum covers many frequencies above and below the visible, which are the colors seen in rainbows.

The second value/definition of color comes from biology the way it is perceived by the brain as mixtures of frequencies.

So according to my logic, it should lie between red and blue in the spectrum. But it doesn't, its wavelength is less than wavelength of blue.

How is this possible?

Because the mixture of frequencies to be perceived as a given color by the brain is seen in the chart. The brain sees the single frequency colors as shown in the first figure, but when frequencies are added new colors and hews are seen.

As the chart is not a simple function it happens that violet does no follow the simplified rule you expect. There is an article in Wikipedia on color vision.

I'd like to challenge your idea that the fact that a mix of red and blue gives something resembling violet implies that violet must be between red and blue.

In practice, what you get from such mixture is a shade of purple. See the figure below. Here the black dashed line from blue to red covers the colors you can get by changing the amount of red and blue in the mixture. The purest violet is on the bottom-most part of the border of the colored shape (the visible gamut).

On the other hand, if you mix violet and sky blue, you can get the set of colors covered by the green dotted line in the diagram. See that blue is among the colors you can get from this mixture.

Do these two facts conflict? No. Most colors are not spectral: they are desaturated. Together they fill a two-dimensional shape, on which the internal points can be found as mixtures of pairs of spectral colors. And the pairs are not unique: e.g. you can get white by mixing orange and sky-blue, or by mixing yellow and royal-blue, etc.. So what you call violet is not a point in this gamut—rather it is an area, where you can pick any point and still call it violet.

You are confused and I understand because even on this site, you can read phrases like "our eyes have cones for Red, Green and Blue light" and "Red light activates the Red cones".

  1. It is a common misconception that the receptors in our eyes are so, that the different types of cones correspond specifically just to Red, Green and Blue light. In reality, the three types of cones are sensitive for a range of Short, Mid and Long (depending on where they are positioned in the visible scale) wavelengths. It is very important to understand that they cover a range of wavelengths, and that they overlap.

Even if you wanted to put a arbitrary conventional color coding for these, you would have to put Yellow, Green and Blue instead of RGB. But these cones have a range of sensitivity, and multiple types of cones could be sensitive for the same wavelength photons. All along the visible wavelength range, there isn't a single position where only one type of cone would be sensitive, that is, every single wavelength photon would activate multiple types of cones.

  1. It is another misconception that whenever monochromatic light shines into our eyes, only one type of cone gets activated. In reality, whenever light shines on our eyes, may it be monochromatic or not, multiple types of cones get activated, it is just the level of activation that is different for different wavelength photons. Whenever monochromatic red light shines into our eyes, even if the photons would be all the same wavelength, they activate both the Long and Mid cones. The more the Long cone gets activated, the more reddish the shade is, the more the Mid range cone gets activated, the more Orange/Yellowish the shade is perceived by our brain.

Our brain is the one that perceives the colors as a combination of signals from the cones, and our brain is the one that perceives violet as a combination of signals from the Short and Long cones too. The more the Short cones get activated, the more violet the shade of the color is perceived by the brain, the more the Long cone gets activated, the more Bluish the shade is perceived by the brain. So this is the answer to your question, violet is the end of the spectrum, where the Short cone's activity dominates. In fact, you need to add the Long cone's activity to get away from the end of the spectrum, and move towards blue.

It is too correct, that a certain perceived color in the brain can be produced by multiple different combinations of signals from the three types of cones.


The Reality of Color Is Perception

P hilosophers have a bad reputation for casting unwarranted doubt on established facts. Little could be more certain than your belief that the cloudless sky, on a summer afternoon, is blue. Yet we may wonder in earnest, is it also blue for the birds who fly up there, who have different eyes from ours? And if you take an object that shares that color—like the flag of the United Nations—and place half in shadow and half in the full sun, one side will be a darker blue. You might ask, what is the real color of the flag? The appearances of colors are frequently changing with the light, and as we move the objects surrounding them. Does that mean that the actual colors change?

All these questions point us to the idea that colors are, despite first appearances, subjective and transitory. Color is one of the longstanding puzzles in philosophy, raising doubts about the truthfulness of our sensory grasp on things, and provoking concerns as to the metaphysical compatibility of scientific, perceptual, and common sense representations of the world. Most philosophers have argued that colors are either real or not real, physical or psychological. The greater challenge is to theorize the subtle way that color stands between our understanding of the physical and the psychological.

My response is to say that colors are not properties of objects (like the U.N. flag) or atmospheres (like the sky) but of perceptual processes—interactions which involve psychological subjects and physical objects. In my view, colors are not properties of things, they are ways that objects appear to us, and at the same time, ways that we perceive certain kinds of objects. This account of color opens up a perspective on the nature of consciousness itself.

LIVING COLOR: In this painting, “The Tree,” by Sudanese artist Ibrahim El-Salahi, the dynamic, wavy black-and-white patterns seem to generate colorful vertical lines. Chirimuuta chose this painting as the cover to her book, Outside Color, because, she says, “I like to think that this symbolizes how color comes into the world because of the continual interactions between perceivers and things perceived.” © 2015 Artists Rights Society (ARS), New York / DACS, London. Image courtesy of Vigo Gallery/Ibrahim El-Salahi

Ingenious: Albert Camus

I had always dreamed of meeting Albert Camus and so was thrilled when he appeared at Lucey’s Lounge, a dark and yellowy lit bar in Brooklyn. The Algerian writer had graciously agreed, or so it seemed, to be interviewed about. READ MORE

For certain philosophers of the ancient world, Greece and India in particular, the variability of perceptual experience from one occasion to the next, and from person to person, raised the worry that the eyes are an unfaithful witness to the world around us. This is because such variability suggests that perceptual experience is as much determined by our own minds as it is by the things we view. Still, colors were not really a problem before the scientific revolution. Discussions of the philosophy of color normally begin their story in the 17th century, at the point at which Galileo, Descartes, Locke, or Newton tell us that sensory, “secondary” qualities—colors, tastes, smells, and sounds—do not belong to the physical world in the way we apparently see that they do.

In The Assayer of 1623, an early Bible of the scientific method and a manifesto for the use of mathematics in understanding the natural world, Galileo writes: “I do not believe that for exciting in us tastes, odours, and sounds there are required in external bodies anything but sizes, shapes, numbers, and slow or fast movement and I think that if ears, tongues, and noses were taken away, shapes and numbers and motions would remain but not odours or tastes or sounds.” 1

Modern science, as inherited from the 17th century, gives us a perspective on material objects that is radically different from our ordinary sensory one. Galileo tells us that the world contains “bodies” which have properties like size, shape, and movement, regardless of anyone perceiving them. By measuring and describing things in terms of those “primary” properties, science promises to give us knowledge of the objective world, the world as it is independently of the distortions of human perception. Science can explain how it is that the molecules released into the air by a sage plant could stimulate my nose, or how its petals could reflect light and appear blue-violet to my eye. But the scent and the color itself—the conscious, sensory experience of them—make no showing in that explanation.

Color is one of the longstanding puzzles in philosophy, raising doubts about the truthfulness of our sensory grasp on things.

The problem of color as we know it today is an ontological issue, a question about what there is in the universe. With the scientific worldview it becomes commonplace to say that the only properties of objects that are unquestionably real are the ones described in physical science. For Galileo they were sizes, shapes, quantities, and motions for physicists today there are more intangible properties like electric charge. This excludes from fundamental ontology any qualitative properties, such as color, that are known to us only through our perceptual faculties. But once colors are excluded, how do we account for their manifest appearance as properties belonging to everyday objects? Either we say that our senses trick us into believing that external objects are colored, when colors do not in fact exist, or we try to find some account of colors that is compatible with a scientific ontology, locating them among material objects.

The view espoused by Galileo has come to be known as subjectivism or anti-realism. The concern is that color perception lands us with an erroneous view of the world and that humans fall victim to a systematic illusion in perceiving external objects as colored. In 1988, the philosopher C.L. Hardin reanimated the Galilean view with the landmark publication Color for Philosophers. 2 His argument drew from the “opponent process theory” of psychologists Leo Hurvich and Dorothea Jameson, which explained color appearances in terms of the brain’s coding of the color signals that come in from the retina. Hardin’s case was that the most adequate account of color must be a neural one. In other words, colored objects are not part of extra-mental physical reality, but a construction or projection of the brain.

Other philosophers have taken up the challenge of finding a place in the material world for these mysterious chromatic qualities. This realism about color comes in many varieties. One proposal is to identify colors with some physical property of objects, like “spectral surface reflectance” (the disposition of surfaces to preferentially absorb and reflect light of different wavelengths). This goes furthest to conserve the common sense notion that colors belong to everyday items in the world, for instance, that the sky is simply and truly blue. The main difficulty is in squaring this with what we know about the subjective aspects of color, like the variability of color as perceivers and contexts change.

SEEING BLUELY: In this photograph of the Blue Mountains near Sydney, Australia, the hills recede into the distance, their appearance becoming more blue and less saturated. Psychologists treat the color as a depth cue, informing the hills’ apparent change in size. To Chirimuuta, the photograph illustrates how perception informs color: “We perceive the distance of the hills in a blue way.” Getty/J.P. Alcarax

The Janus-Facedness of Color

The problem with these realist and anti-realist proposals is that they each only focus on either the objective or subjective aspects of color. An alternative position can best be described as “relationist.” Colors are analyzed as perceiver-related, but nonetheless real properties of objects. The account is salient in 17th-century literature (notably John Locke’s Essay Concerning Human Understanding), encapsulated in the idea that colors are dispositions of objects to appear in a certain way. It is interesting that this relationist proposal fits in with some current ideas in the science of color perception. Vision scientists Rainer Mausfeld, Reinhard Niederée, and K. Dieter Heyer write that, “the concept of human color vision involves both a subjective component, as it refers to a perceptual phenomenon and an objective one . We take this subtle tension to be the essential ingredient of research on color perception.” 3

Later in the same article they call this quality the “Janus-facedness” of color: Color points out to the world of objects, and at the same time it draws us inward to examine the perceptual subject. This is a common thread in scientific writing on color vision and it has always struck me that the Janus-facedness of color is its most beguiling quality.

Instead of treating color words as adjectives, we should treat them as adverbs. I eat hurriedly, walk gracelessly, and I see the sky bluely!

In an influential textbook, perceptual psychologist Stephen E. Palmer writes that color is not reducible to visual experience or properties of objects or lights rather, Palmer writes, “Color is more accurately understood as the result of complex interactions between physical light in the environment and our visual nervous systems.” 4

Indeed, I argue, colors are not properties of minds (visual experiences), objects or lights, but of perceptual processes—interactions that involve all three terms. According to this theory, which I call “color adverbialism,” colors are not properties of things, as they first appear. Instead, colors are ways that stimuli appear to certain kinds of individuals, and at the same time, ways that individuals perceive certain kinds of stimuli. The “adverbialism” comes in because colors are said to be properties of processes rather than things. So instead of treating color words as adjectives (which describe things), we should treat them as adverbs (which describe activities). I eat hurriedly, walk gracelessly, and on a fine day I see the sky bluely!

It is common for physicists to explain the blue appearance of the sky as due to “Rayleigh scattering,” the fact that short wavelengths of visible light are scattered more by the Earth’s atmosphere than longer ones, so that diffuse blue light comes to us from all regions of the sky when the sun is high and cloudless. But we should not be tempted to say the blue of the sky is simply a property of the scattered light. There is no blueness unless the light interacts with perceivers like us, who have photoreceptors that respond differently to short versus long wavelengths of light.

So, precisely speaking, the sky is not blue. We see it in a blue way.

For the adverbialist, there is no color-in-the-object on the one hand, and color-in-the-mind on the other. Color is the property of a perceptual process. Because color cannot be reduced either to physics or to psychology, we are left with a blue sky that is not simply inner or outer but somehow in between.

This idea has implications for the understanding of conscious perceptual experience. We’re used to thinking of conscious experience as something like a series of sounds and images rolling past on an inner movie screen. This is the conception of our mental life that the philosopher Alva Noë wants to break away from. In his 2009 book Out of Our Heads, Noe claims that consciousness is not confined to the brain but is somehow “in between” the mind and our ordinary physical surroundings, and that consciousness must be understood in terms of activities. 5 By themselves these ideas are quite perplexing. But taking the example of visual experience, color adverbialism is a way to make sense of consciousness being “out of our heads.” According to adverbialism, color experience comes about because of our interaction with the world, and would not exist without this exposure to our surroundings. Our inner mental lives are dependent on this outer context.

Ultimately, the philosophical tool of color adverbialism suggests a new way to get out of the traditional internalist conception of the mind, making vivid the bridge between our mental lives and the outer world.

Mazviita Chirimuuta is an assistant professor in history and philosophy of science at the University of Pittsburgh. Her book Outside Color has recently been published by MIT Press.

1. Galileo, G. The Assayer in Drake, S. Discoveries and Opinions of Galileo Knopf Doubleday Publishing Group, New York, NY (1957).


The Human Eye and RGB Colour

Things have colour because light has properties that are visible to the human eye. But what does this mean?

  • The colour of objects depends firstly on the light source and the wavelengths it emits.
  • The way any object appears to an observer depends on the material it is made from and what happens when light strikes (or is emitted from) its surface.
  • Light striking an object may undergo absorption, dispersion, reflection, refraction, scattering or transmission.
  • The appearance of an object to an observer also depends on the mental processes that lead to colour perception.
  • Spectral colours are produced by a single wavelength of light or by a band of similar wavelengths. is an additive colour model in which red, green and blue light is added together in various proportions to reproduce a wide range of other colours. The name of the model comes from the initials of the three additive primary colours &ndash red, green, and blue.

Remember that:

When light strikes an object at least one of the following things happens:

  • Absorption. When light strikes an opaque medium the wavelengths that are not reflected are absorbed and their energy is converted to heat.
  • Dispersion. Chromatic dispersion refers to the way that light separates into its component wavelengths and the colours corresponding with each wavelength become visible.
  • Reflection. The term reflection refers to a situation where light strikes the surface of an object and some wavelengths are obstructed and bounce off. If the surface is smooth, light is reflected away at the same angle as it hits the surface. The term reflection refers then to what happens to wavelengths of light that are neither absorbed (by an opaque medium) nor transmitted (through a transparent medium).
  • Refraction. The term refraction refers to the way a light wave changes direction and speed as it travels from one medium to another.
  • Scattering. Scattering takes place when light waves are reflected in random directions at the boundary between two media. Scattering can also take place when light strikes particles or other irregularities within a medium through which light propagates.
  • Transmission. In optics, transmission refers to the passage of electromagnetic radiation through a medium.

Follow the blue links for definitions . . . . or check the summaries of key terms below!


Materials and methods

Animals

Twelve female quail Coturnix coturnix japonica (Linnaeus) were kept in groups of six, each in a floor pen that measured 2.6m×1.8m. Six individuals had been raised in UV+ conditions, while six individuals had been raised in UV- conditions. The UV+ conditions consisted of full-spectrum fluorescent lamps [Durotest Truelite, for irradiance spectra, see Hunt et al., 2001, running on high-frequency ballasts (>30 kHz Cooper Lighting and Security Ltd,Doncaster, UK)]. These tubes are designed to mimic natural sunlight in their approximate balance of UV and longer human-visible wavelengths(Bennett et al., 1996). The UV-lighting conditions were created by covering these lamps with a UV-blocking filter (Lee 226 UV- filter, Lee filters, Andover, UK see Fig. 1 for transmission spectra). Two of the quail reared in UV+ conditions and two of the quail reared in UV- conditions were tested when they were between four and eight months of age. Four European starlings Sturnus vulgaris (Linnaeus)were wild-caught as juveniles under an English Nature licence (#20000069) in Somerset, UK and were maintained under UV+ conditions in the laboratory. The starlings were tested in our experiments when they were between six and eight months of age.

Transmission spectra of the three filter types used in the three experiments. Lee UV- (lights) refers to the flexible Lee 226 UV-blocking filter used to remove UV wavelengths from the ambient light by covering the light sources this occurred in probe trial 2 of experiments 1, 2 and 3. UV+(patterns) and UV- (patterns) refer to the UV-transmitting and UV-blocking solid Perspex filters used to cover the training stimuli, respectively.

Transmission spectra of the three filter types used in the three experiments. Lee UV- (lights) refers to the flexible Lee 226 UV-blocking filter used to remove UV wavelengths from the ambient light by covering the light sources this occurred in probe trial 2 of experiments 1, 2 and 3. UV+(patterns) and UV- (patterns) refer to the UV-transmitting and UV-blocking solid Perspex filters used to cover the training stimuli, respectively.

Stimuli

Perceptual ability was tested by giving the birds a discrimination task in which they were allowed to move freely around a foraging arena. In this arena,there were always eight stimuli that overlay separate food wells (1.5 cm×1.0 cm diameter × depth). In every trial, four stimuli of one colour were rewarded, and four stimuli of another were not rewarded with food. If birds can perceive and remember the difference between the two sorts of stimuli, then they should learn to ignore the unrewarded stimuli.

The stimuli were 2.5 cm×2.5 cm patterns consisting of a tiling of 121 grey squares of varying intensity (see Fig. 2 for examples). Birds are able to resolve at least four cycles per degree (Schmid and Wildsoet,1997), so the discrete squares should have been perceptible to the animals. Each pattern was attached to the upper surface of a 37 g metal weight of the same size as the pattern (Fig. 2) to increase the energetic cost of moving the stimuli and thereby promote learning. On their lower surfaces, each weight was completely coated with matt black paint. The sides and bottoms of all the weights were laminated with Sellotape to prevent chipping of the paint, which may have provided the birds with alternative cues with which to solve the task during training. Birds were trained on three different visual discriminations, which generated the three different experiments described below. In each experiment,there were 12 pairs of training patterns.


If longer wavelength color is perceived faster by human, what about non-spectral color (black)? - Biology

A study uses sugar water experiments to show that hummingbirds can see colors invisible to us.

  • Hummingbirds can see colors in the ultraviolet range. We cannot.
  • The tiny powerhouse derive hues from four types of photoreceptors, as opposed to our three.
  • As beautiful as the world already is, let's talk about what hummers see.

Do you know what the first three nonspectral colors are? Nope. Neither do we or any other humans. Maybe you should ask a hummingbird. The colors we know are part of the visible spectrum, a series of electromagnetic waves whose lengths are between 380 and 700 nanometers. (Electromagnetic waves repeat, and a wavelength is the distance between one repeat and the next.)

There are electromagnetic waves whose length is shorter or longer, but we lack the ability to see them. There is some evidence that other species can see these wavelengths — most famously the mantis shrimp — and a new study affirms that hummingbirds can indeed detect non-spectral wavelengths, and thus colors. As if these little creatures weren't already amazing enough.

Incredible already

People all over the world exhibit a fascination with the tiny, beautiful avians that are birds unlike any other. Minute yet powerful, these incredible little energy bombs annually migrate great distances, living only on nectar and sugar water during visits to gardens — anyone who's ever annoyed a hummer by changing their food at feeding time can attest that these little buzzing creatures are not shy. Before getting into hummingbirds' remarkable color perception, here are a few other jaw-dropping stats:

  • Hummingbirds can beat their wings from 20 to 200 times per second.
  • If a human were to burn as many calories as a hummingbird, that human would have to consume 155,000 calories per day.
  • Hummingbirds have larger hearts (by body size) than any other bird, which is good because it can beat 12,000 times per minute.
  • The dazzling, iridescent red neck of a male ruby-throated hummingbird is an optical illusion — their chest is actually black, brown, and reddish brown.

Image source: Steve Byland/Shutterstock

So many colors

The new study in Proceedings of the National Academy of Sciences of the United States of America explains why its authors believe hummingbirds can see non-spectral colors.

First, the researchers recruited some volunteers: wild broad-tailed hummingbirds (Selasphorus platycercus). The experiments took place at the Rocky Mountain Biological Laboratory (RMBL) in Colorado, an environment researchers took pains to keep as natural as possible during the three-year study. Scientists from Princeton, the University of British Columbia (UBC), Harvard University, University of Maryland, and RMBL were involved in the experiments.

Before dawn each day, the researchers set up a pair of feeders for the birds — one with a rewarding drink of sugar water, the other with unsweetened plain water. Next to each feeder was an LED tube capable of emitting a broad range of colors, including nonspectral colors. Over the course of several hours, the hummingbirds learned that one color — sometimes a nonspectral color — signified the rewarding tube, and another color the plain water. When the feeders' positions were swapped, the hummingbirds simply followed the color, even if it was one the researchers themselves couldn't discern.

Study co-author Harold Eyster recalls, "It was amazing to watch. The ultraviolet+green light and green light looked identical to us, but the hummingbirds kept correctly choosing the ultraviolet+green light associated with sugar water. Our experiments enabled us to get a sneak peek into what the world looks like to a hummingbird."

The tests were run with various spectral and nonspectral color pairings for the two feeders, and the hummingbirds apparently couldn't have cared less which kind of color was employed — they quickly learned where the sugar-water feeder was. The researchers also ran control experiments to make sure the birds weren't being tipped off by smell or some other cue.

Though the experiment was fundamentally pretty simple, the results are stunning. Lead author Mary Caswell Stoddard of Princeton says, "To imagine an extra dimension of color vision — that is the thrill and challenge of studying how avian perception works."

How hummingbirds do this

Human eyes have three types of color receptors, or cones, each of which responds most strongly to a specific range of wavelengths — S-cones specialize in blue, M-cones green, and L-cones red. From combinations of those three basic hues, our eyes and brains present us the millions of colors we perceive.

Though hummingbirds come nowhere near the mantis shrimp's twelve-plus collection of cone types, they do have four, which endows them with tetrachromacy. If we, with our trichromacy, can construct so many colors from three basic hues, imagine what adding a fourth might do. In the hummingbirds' case, the fourth cone type perceives ultraviolet light that can be added to the other three hues for unimaginable (to us) combinations.

"Humans are color-blind compared to birds and many other animals," points out Stoddard. We can only wonder what these colors actually look like to hummingbirds. As RMBL's David Inouye says, "The colors that we see in the fields of wildflowers at our study site, the wildflower capital of Colorado, are stunning to us, but just imagine what those flowers look like to birds with that extra sensory dimension."


Download Agreement

Before downloading or cutting and pasting from lightcolourvision.org we ask you to agree to the following terms:

  1. All information, imagesand other assets displayed or made available for download on the lightcolourvision.org website are copyright. This means there are limitations on how they can be used.
  2. All information, imagesand other assets displayed or made available for download are solely and exclusively to be used for personal, educational and non-profit purposes.
  3. When you find the resources you need, then part of the download process involves you (the user) ticking a box to let us (at lightcolourvision.org) know we both agree how the material can be used.
  4. Please contact [email protected] before considering any use not covered by the terms of the agreement above.

If you are a student, educator or researcher you might also like to read our posts Copyright: Credit where credit&rsquos due and Citation and bibliographies before copying and pasting material into essays, assignments or other academic work. They include advice on avoiding plagiarism and how to credit other people&rsquos words, images and assets before submitting your work for marking or assessment. If you are confused, just ask a friendly teacher, librarian, or other member of academic staff.

Like to know more?

Please contact [email protected] if you have questions about any aspect of this project.

The normal is an imaginary line drawn on a ray diagram perpendicular to, so at a right angle (90 degrees), to the boundary between two media.

Yes! Refraction takes place as light crosses the boundary between one transparent medium and another with a different refractive index.

Yes! Refraction takes place as light crosses the boundary between one transparent medium and another with a different refractive index.

A human observer is a person who watches something from their own unique point of view.

Yes! Light becomes diffused as it is transmitted through a medium or is reflected off its surface because of the effect of the scattering of light.

Total internal reflection means that all the light propagating through a medium is reflected when it reaches its boundary.

The reflectance of a surface refers to its effectiveness at reflecting light.

In the field of optics, diffusion refers to the scattering of light as a result of reflection or transmission.

No! Human vision relies on trichromacy which is not the same as the RGB colour system.

However, when red, green and blue light is mixed together in different proportions the human eye sees all the colours of the visible spectrum.

Input a search term in the form above:

| Additive | Angle | | Cell | Chromatic dispersion | CMY | CMYK | Colour | Colour constancy | Colour | Colour model | Colour wheel | Continuous spectrum | Crest | Dispersion | Electric field | Electromagnetic radiation | Electromagnetic spectrum | Electromagnetic wave | Energy | Eye | Frequency | Gamma ray | Hertz | Incident light | Index of refraction | Law of refraction | Light | Long wave | Low frequency | Media | Medium | Monochrome | Non-spectral | Normal | Optical | | Optics | Primary colour | Propagate | Radio wave | Rainbow | Rainbow colour | Ray | Reflection | Refraction | Refractive index | RGB colour | ROYGBV | RYB | Secondary colour | Short wave | Sine | Snell’s law | Spectral colour | Speed of light | Subtractive colour | Tertiary colour | Transparent media | Vacuum | Visible light | Visible spectrum | Vision | Wave | Wavelength | White light |

A non-spectral colour is any colour that is not be produced by a single wavelength of visible light.

Yes! The wavelength and speed of light a ray of light change as they travel through different media.

The refractive index of a medium is calculated by dividing the speed of light in a vacuum by the speed of light as it propagates through another medium.

So the refractive index of a vacuum = 1 because the speed of light in a vacuum divided by the speed of light in a vacuum = 1.

The angle between the electric and magnetic field is 90 degrees.

Infrared and ultraviolet are forms of electromagnetic radiation with wavelengths just outside the visible spectrum.

The unit used to measure wavelength is the metre.

Because the size of electromagnetic waves varies, different prefixes are used to aid measurement. Here are six examples: kilometre, centimetre, millimetre, micrometre, nanometre and picometre.

The frequency of incident light is unchanged as it travels from air into glass so its colour remains the same.

The unit used to measure the frequency of light is the hertz.

The frequency of visible light corresponds to a band of frequencies in the vicinity of 430–770 terahertz (THz)! There are a trillion (1,000,000,000,000) terahertz in a hertz!

Yes! As the frequency of oscillations of an electromagnetic wave increases the wavelength decreases.

Yes! Energy increases with frequency.

Violet has the highest frequency of visible light.

There are 1000 Kilohertz in one Hertz.

The hertz is used to measure the frequency of electromagnetic waves!

Because the frequency of electromagnetic waves varies enormously, different prefixes are used to aid measurement. Here are four examples: Kilohertz, megahertz, gigahertz and terahertz!

There are trillion (1,000,000,000,000) picometres in a metre.

Yes! Gamma rays have the shortest wavelengths within the electromagnetic spectrum.

Gamma rays have a shorter wavelength than radio waves.

Yes! Waves with lower frequencies have longer wavelengths.

The unit used to measure the wavelengths of visible light is the nanometre.

In the case of light waves, to propagate means to travels through a medium in a particular direction.

An electromagnetic wave is the result of the interaction of electric and magnetic fields.

Yes! A wave with lower frequency has a longer wavelength.

Gamma rays transport more energy than any other band of wavelengths within the electromagnetic spectrum.

Gamma rays have the shortest wavelengths of any type of radiation within the electromagnetic spectrum.

The unit used to measure wavelength is the metre. Because the size of electromagnetic waves varies, different prefixes are used to aid measurement. Here are six examples: kilometre, centimetre, millimetre, micrometre, nanometre and picometre.

The range of wavelengths that correspond with green is between 570 – 495 nanometres.

Red has the longest wavelength whilst violet has the shortest.

Yes! Wavelength is measured in nanometres.

Yes! The colour of visible light depends on its wavelength.

Yes! A light wave is another name for an electromagnetic wave! But remember that not all forms of light are visible to the human eye.

When white light strikes an object, its colour is determined by which wavelengths of light are absorbed and which wavelengths are reflected towards the observer.

The ball is painted cyan! (Now check out the next question).

Sunshine is important to human beings because without light we can’t see.

Sunlight can be described in terms of both waves and particles.

Too much ultra-violet radiation causes sunburn.

Yes! The energy that the sun emits is called electromagnetic radiation.

No! Light only becomes visible when it strikes a medium or object.

No! The visible spectrum is just the small part of the electromagnetic spectrum our eyes are tuned to.

Because different wavelengths of light are reflected off the surface of objects. Every surface has unique properties.

Wavelengths visible to the eye are in a band between approximately 390 to 700 nanometres.

Yes! Visible light is a form of electromagnetic radiation.

Red, orange, yellow, green, blue and violet are the bands of colour we see in a rainbow.

Red is the band of colour with the longest wavelength.

Violet is the band of colour with the shortest wavelength.

nm is shorthand for nanometre.

Rainbows produce spectral colours as sunlight is refracted by raindrops.

The invisible band of wavelengths next to red is infrared.

Yes! Each wavelength of light corresponds with a different colour seen by an observer.

The unit of measurement for wavelengths of visible light is the nanometre (nm).

When wavelengths of light corresponding with red, green and blue are projected at equal intensity onto a dark surface it appears white to an observer.

White light contains all the wavelengths of the visible spectrum at equal intensity.

Yes! Because the Sun radiates light at all wavelengths of the electromagnetic spectrum it therefore also emits light at all wavelengths of the visible spectrum.

The Sun radiates light at all wavelengths of the electromagnetic spectrum.

Yes! As the frequency of oscillations of an electromagnetic wave increase the wavelength decreases.

Yes! Energy increases with frequency.

Hertz. Kilohertz. Megahertz. Gigahertz. Terahertz.

A light wave in a vacuum travels at 300,000 kilometres (km) per second! Or to be exact, 299,792 km/sec.

Yes! A wave-cycle can be measured from any point on a wave to the same point on the next wave.

The lowest point in the oscillation of a wave is called the trough.

Yes! light waves propagate through a vacuum in a straight line.

Blue and violet are two of the colours with the shortest wavelengths.

Red and orange are two of the colours with the longest wavelengths.

Some estimates of the number of colours the human eye can distinguish between run into the millions.

Yes! Every wavelength of light corresponds with a different colour.

Our eyes are tuned to visible light – the visible part of the electromagnetic spectrum.

Names for solar radiation include solar energy and light.

The sun emits electromagnetic radiation.

The sun generates energy as a result of thermonuclear fusion.

The star at the centre of our solar system is called the Sun.

Gamma rays have a higher frequency than any type of electromagnetic radiation.

Gamma rays transport more energy than any other form of electromagnetic radiation.

Shorter wavelengths = Higher frequency.

Lower frequency = Longer wavelengths.

Yes! The speed of light is affected by the medium through which it propagates.

The maximum speed of light occurs in a vacuum! Light travels in air at 99% of the speed of light in a vacuum.

No! Both crown glass and diamonds are slow media because they significantly reduce the speed of light.

No! The colour of a ray of light remains the same because frequency doesn’t change as light travels through different media.

Millimetres, centimetres, metres and kilometres are all used to measure the wavelengths of radio waves.

RGB refers to the colours red, green and blue. These are the primary colours used by the RGB colour model to mix wavelengths of light to produce a palette of as many as 16 million colours.

ROYGBV refers to red, orange, yellow, green, blue and violet. ROYGBV are the spectral colours associated with rainbows and the diffusion of white light.

The visible spectrum is the small part of the electromagnetic spectrum our eyes are tuned to.

White light is the name for light containing all the wavelengths of the visible spectrum.

A continuous spectrum is produced by an inclusive band of wavelengths of light between any two points on the electromagnetic spectrum.

Additive primary colours are three wavelengths of light that produce white when combined together in equal proportions.

Red, green and blue are the three additive primary colours used in the RGB colour model.

Yes! Every spectral colour corresponds with a single wavelength of visible light.

The visible spectrum includes all the wavelengths of light the human eye is sensitive to and results in the colours we see between red and violet.

Additive primary colours are three wavelengths of light that produce white when combined together in equal proportions.

Spectral colours are all the colours between red and violet that can be produced by a single wavelength of light. Sunlight is composed of spectral colours.

A typical human eye will respond to wavelengths between 390 to 700 nanometers.

Yes! Each colour in a rainbow between red and violet is a spectral colour.

No! The colours produced by mixing RGB primary colours are not spectral colours because they are not produced by a single wavelength of light.

RGB is a colour model used to produce a full palette of colours by mixing red, green and blue light sources in different proportions.

Secondary colours are the colours produced by mixing pairs of primary colours in equal proportions. The RGB secondary colours are cyan, magenta and yellow.

Yes! ROYGBV are all spectral colours and so each can be produced by a single wavelength of light between 480 and 700 nanometres.

Yes! ROYGBV are all spectral colours and so can be produced by a single wavelength of light.

Yes! Spectral colours can be combined together to produce orange, yellow and violet in the correct proportions.

Cone cells are the light-sensitive neurons in the retina at the back of our eyeballs.

When red (660 nm), green (525 nm) and blue (460 nm) colours of light are projected at the same intensity onto a neutral coloured surface they produce white.

Cyan, magenta and yellow are the three primary colours used for digital printing. They are subtractive primaries, so mixed together they produce black.

When wavelengths corresponding with red, green and blue are projected onto a neutral coloured surface they produce white.

Red, green and blue are the three primary colours in the RGB colour model.

There are 6 tertiary colours in the RGB colour model! They result from mixing a primary and secondary colour. So a tertiary colour is produced by mixing red-yellow or green-cyan etc.

An RGB colour wheel is a way of exploring the relationship between red, green and blue primary colours when they are mixed to produce secondary and tertiary colours etc.

Spectral colours are the colours produced by different wavelengths of light. Sunlight and rainbows are composed of spectral colours. RGB colour is an additive colour model in which red, green and blue light is combined in various proportions to produce the appearance of other colours.

Red and green are the two primary colours of light that together make yellow.

When spectral colours are arranged in a diagram, the order in which they appear corresponds with their wavelength and so their place in the visible part of the electromagnetic spectrum.

A computer screen uses the RGB colour model. Each pixel contains three tiny semiconductors that produce red, green and blue light! The colour we see changes as the intensity of light produced by each semiconductor increases or decreases!

Because a colour wheel demonstrates the effect of mixing different proportions of the three RGB primary colours – red, green and blue!

Cyan is a secondary colour in the RGB colour model!

In the RGB colour model green and blue are the two primary colours that together make cyan!

Yes! Cyan is a spectral colour with a wavelength of around 510 nanometres (nm).

Magenta is a secondary colour in the RGB colour model.

A printer that uses CMYK inks adds black (K) ink to cyan, magenta and yellow to produce deeper blacks.

No! There is no single wavelength in the visible spectrum that corresponds with magenta.

Cyan is a primary colour in the CMY colour model.

Blue and red are the two primary RGB colours that together make magenta!

When wavelengths corresponding with red, green and blue are projected onto a neutral coloured surface in equal proportions they produce white.

Inks corresponding with the three CMY primary colours produce black during printing because each colour subtracts from the wavelengths of light reflected off the paper towards an observer.

Overlapping wavelengths of light corresponding with cyan, magenta and yellow make white because, when reflected off a neutral coloured surface, each adds more wavelengths to the reflected light.

There are three primary colours in an RGB colour wheel.

There are twelve tertiary colours in an RGB colour wheel with eighteen colours.

Yes! All RGB colour wheels start with the three primary colours: red green and blue.

Yes! There are always secondary colours between primary colours.

No! Because of the way colour wheels divide up when using decimal or hexadecimal notation, the intermediary colours between secondaries do not always include tertiary colours. Tertiary colours are produced by using equal proportions of secondary colours.

Yes! Every colour on a colour wheel is produced by mixing equal proportions of the colours on either side.

Red, green and blue are the three primary colours used when building an RGB colour wheel.

Cyan, magenta and yellow are the three secondary colours when building an RGB colour wheel.

Yes! Equal proportions of primary colours are used to produce secondary colours.

Yes! Every colour in a colour wheel can be produced by mixing different proportions of red, green and blue primary colours.

The wavelength of an electromagnetic wave is a measurement of the length of a single oscillation of the wave.

The frequency of an electromagnetic wave is a measurement of the number of wave oscillations passing a given point in a given period of time.

Frequency of an electromagnetic wave in a vacuum is calculated by dividing the speed of light by the wavelength of the wave.

Yes! The speed of light depends on the optical density of the medium it is propagating through.


Color Vision by Peter Gouras

Color vision is an illusion created by the interactions of billions of neurons in our brain. There is no color in the external world it is created by neural programs and projected onto the outer world we see. It is intimately linked to the perception of form where color facilitates detecting borders of objects (Figure 1).

Color is created by utilizing two properties of light, energy and frequency of vibration or wavelength. How our brain separates these two properties of light, energy and wavelength, and then recombines them into color perception is a mystery that has intrigued scientists through the ages. We know much about the nature of light and the subjective impressions of color, definable by physical standards (Wright, 1946) but ultimately color should be explained at the level of single cells in our brain. Examination of the responses of single neurons or arrays of such neurons provides the best insights into the physiology of color vision. Ultimately our understanding of this process will allow us to model the neural circuits that underlie the perception of color and form. Although still beyond reach, progress is being made in deciphering these clever circuits that create our perception of the external world.

We start by describing the nature of the photoreceptors that convert light energy into neural signals. Then we consider the parallel channels leading from the retina to the thalamus carrying information into visual cortex, where color is ultimately determined. Lastly we use our current understanding to speculate on how visual cortex uses neural circuits to create the perception of color and form.

2. The Photoreceptors.

Photoreceptors are neurons specialized to detect light. The detection occurs in an organelle called the outer segment, a membranous structure where light absorbing proteins, opsins, are embedded. There are two major types of photoreceptors in most vertebrate eyes, rods and cones. Rods are very sensitive but slow and their response saturates at light levels where cones function optimally. Rods are are not used much in modern society where artificial illumination adequate for cone vision is ubiquitous. Cones are less sensitive but are fast and can adapt to the brightest lights, being almost impossible to saturate. Cones evolved before rods undoubtedly in areas of strong sunlight where vision was a great advantage. In broad sunlight, shadows are strong and more important to detect than increments of light in the struggle for survival. Shadows depolarize cones leading to a release of a transmitter that influences second order retinal neurons. The appearance of light hyperpolarizes cones leading to a curtailing of this transmitter.

A cone responds only to the energy it absorbs (Maxwell 1872). All wavelength of light can produce identical responses from a cone if the energy absorbed by the cone is the same for these wavelengths (Figure 2). Cones are therefore color blind producing a univariant response reflecting only the amount of energy they absorb. Detecting objects by the energy reflected from their surfaces, however, can fail when objects reflect a similar amount of energy as their background. Here is where color vision becomes important. Wavelength contrast can detect objects when energy contrast is absent or minimal. An object can reflect the same energy but seldom reflects the same wavelength composition as its background. Color vision combines both energy and wavelength contrasts to detect objects and this advantage must have evolved early in the evolution of vision.

In order to detect objects by differences in spectral reflectance, two or more different types of cones are needed. This is an important concept for understanding color vision. For divariant color vision, two cone types must exist and be sensitive to different parts of the visible spectrum, preferably as different as possible. The range of the visible spectrum depends on the ability of light to penetrate the eye and be absorbed by the photoreceptors. Ultra-violet light is absorbed by the anterior segment of our eyes and seldom reaches the photoreceptors. Infra-red light penetrates our eye readily but its quantal energy may be too small to activate opsins. Therefore early in the evolution of color vision, opsins sensitive to the middle of our visible spectrum evolved, near spectral yellow, and a short wavelength opsin evolved in a second type of cone, near spectral blue (Figure 3). These have been called L (long wavelength sensitive) and S (short wavelength sensitive) cones, respectively, and this was a first step in the evolution of color vision.

In animals with large eyes, an interesting strategy evolved. L-cones were used to detect both energy and wavelength contrast but S-cones were used only for wavelength contrast. This was due to chromatic aberration. Short wavelength images are out of focus when longer wavelength images are in focus on the photoreceptor mosaic. Chromatic aberration increases greatly at short wavelengths, which led to the L cone system dominating energy contrast. As a result there are many more L than S cones in many mammals in order to gain spatial resolution by achromatic contrast detectable by L cones.

In animals with small eyes, like mice and rats, ultra-violet light can reach the photoreceptor mosaic and in this case ultra-violet sensitive cone opsins have evolved to widen the spectral range of vision and if combined with L cones could allow color vision (Haverkamp et al., 2005 Ekesten and Gouras, 2005 Yin et al., 2008). Chromatic aberration is reduced in these small, highly spherical eyes which have outer segments as long as animals with large eyes. This increases their depth of focus minimizing chromatic aberration, an advantage of being small. Their retinal images are less magnified, however, than those in large eyes.

3. Chromatic and Achromatic Contrast

In order to establish chromatic contrast it is necessary to compare the responses of a group of cones of one type with the responses of a group of cones of another type within the same area of visual space. One cannot compare only two neighboring cones such as the S and the L cone in Figure 4A.

If an image of gray and yellow were to cover these two cones in an inappropriate way as done in Figure 4A the brain would reach an erroneous conclusion about this border because the S cone would not be affected by the yellow side and the L cone would be strongly affected by the gray side. The brain might consider this to be a black/yellow border rather than a yellow/gray border. Even if a slit of light imaged on only the more numerous L cones (Figure 4B), there would be ambiguity about the colors of adjacent images. If only L cones are affected by both yellow and gray light, the brain would not see any border here. Only when a larger image which covers a number of S and L cones (Figure 4C) can an unequivocal decision be made about the color of these bordering images. In this case the left side, which is yellow, would strongly affect L but not S cones and therefore be judged to be yellow. The right side of this border, which is gray, would strongly affect both L and S cones and therefore would appear white or gray. For achromatic contrast, smaller images can be analyzed for border contrasts and in most cases only two neighboring cones would be sufficient to distinguish a light/dark border. This is why a unit area of achromatic space is smaller than a unit area of chromatic space.

4. Horizontal Cells.

Cones receive an antagonistic input from horizontal cells, whose cell bodies reside in the outer nuclear layer and whose processes contact the spherules of rods and the pedicles of cones (Kolb, 1991). There are at least two varieties of cone horizontal cells (Figure 5).

One variety (H1) only contacts L cones the other variety contacts both S and L cones (H2). Cone horizontal cells receive an excitatory input from cones and send back an inhibitory input to cones. This is a type of negative feedback. When a cone is hyperpolarized by an increment or depolarized by a decrement of light, it receives an opposing input from horizontal cells after a brief synaptic delay. This dampens the response and can also reduce the effects of scattered light by minimizing cone responses outside of the focal image on the retina. In color vision the horizontal feedback also acts to narrow the action spectrum of cone bipolar cells. In divariant color vision this can narrow the action spectrum of S cone bipolar cells. This occurs because the processes of the H2 horizontal cell that reach L cones are only post-synaptic to S cones. Therefore L cones can send an antagonistic signal to S cones which can reduce the effectiveness of wavelengths, absorbed by both L and S cones and this narrows the action spectrum of the S cone channels (Packer et al., 2007).

5. Bipolar and Ganglion Cells.

The retina is composed of three layers of neurons, the outermost being the photoreceptors, rods and cones, which for divariant color vision are L and S cones. A second layer of bipolar cells transmits the signals of the photoreceptors to a third layer of neurons, the ganglion cells whose axons form the optic nerve. The ganglion cells serving the two photoreceptor systems are quite different. The L cones synapse with unique bipolar cells, called “midget” bipolar. These single cone detectors were discovered by Stephan Polyak in monkey retina using the Golgi method of silver impregnation. Because of their small size they were called “midget” bipolar cells. This provides the brain with the ultimate in spatial resolution, a single cone, and in addition isolates the signals of L cones which can be used for color vision. A feature of this L cone bipolar system is the presence of two types of cells (Figure 6). One type of L cone bipolars depolarizes whenever the cone or cones they synapse with hyperpolarize this set is called on-bipolars because they are excited (turned on) by light. The other set is depolarized whenever the cones they synapse with detect decrements of light this set is called off-bipolars because they are excited by darkness and inhibited (turned off) by light. These two sets of cone bipolar cells synapse with separate sets of on and off ganglion cells at two levels in the inner plexiform layer of the retina, a more external off-lamina and a more internal on-lamina (Nelson et al., 1978).

These parallel channels, transmitting lightness and darkness from local retinal areas, are maintained throughout the visual pathway to visual cortex. This “push-pull” system of neurons is thought to increase the dynamic range for detecting decrements and increments of light in local areas of the retinal image. Only L cones seem to be connected to both on and off bipolar cells while S cones are connected to only on-bipolar cells. The reason for this may be that the latter are only involved in chromatic vision while the former are involved in both chromatic and achromatic vision. Achromatic vision involves the detection of lightness and darkness while chromatic vision involves the detection of color. Not only do S cones lack an off-bipolar system but they have a much different route to the ganglion cell output layer. The S cone on bipolar excite the internal arbor of a bistratified ganglion cell while a wide-field L cone off bipolar excites the external arbor of this same ganglion cell (Dacey & Lee, 1994). This S cone system is absent in the very center of the fovea where achromatic contrast is mediated by the L cone midget system. Away from the fovea the L cone midget system begins to contact more than one cone and therefore loses spatial resolution.

There is a second ganglion cell system that plays no role in color vision but is also connected to only L cones. These are parasol ganglion cells. They are larger cells with faster conduction velocities and they target the magno-cellular layers of the lateral geniculate nucleus (LGN). They appear to play a role in the detection of movement and possibly slow tracking movements. There is some evidence that they might receive an input from S cones but I have never been able to detect such an input. Barry Lee’s group considers these cells play a role in the detection of luminance (Lee, 2008).

6. Divariant Blue/Yellow Color Vision

The ganglion cells in Figure 6 are thought by most to be the essential channels for divariant color vision. The bistratifid S cone on-cells and the L cone midget cells monitor the light absorption of S and L cones respectively. The signals from these two channels must be compared in the same areas of visual space for color vision. The cone bipolar cells driving the bistratified S cone on-ganglion cells have co-extensive fields (Figure 7), which is ideal for comparing differences between two cone mechanisms in the same area of visual space.

These two different bipolar inputs to the bistratified ganglion cell do not appear to oppose each other because these cells are excited by white light. Another more indirect input to these cells comes from the H2 horizontal cell which transmits antagonistic signals from L to S cones. This input provides spectral antagonism to the bistratified ganglion cell because light activation of L cones will produce a depolarizing signal in S cones that counteracts the hyperpolarization produced by short wavelength light. The strong response to white light, however, implies that this H2 mediated antagonism is relatively weak.

The channel transmitting L cone signals for spectral contrast in visual cortex has been considered to be the midget system, at least in primates. These midget ganglion cells are considered to receive no input from S cones, either synergistic or antagonistic, although there are connections to S cones through H2 horizontal cells. In trivariant monkey retina there is no evidence of S cone input to either the midget or the parasol ganglion cell systems which implies that H2 horizontal cells are only post- and not pre- synaptic to L cones. The H1 horizontal cell only contacts L cones and therefore provides only spatial antagonism to neighboring L cones and does not produce spectral antagonism.

This previous picture is not accepted by everyone. There is evidence that a L cone on/S cone off ganglion cell exists to provide an opponent L cone on input to visual cortex (Tailby, 2008: Neitz and Neitz, 2008 Martin et al., 1997 Chatterjee and Callaway, 2003). It has been difficult to find such a ganglion cell in primate retina but they they have been reported in the koniocellular layers of the primate lateral geniculate nucleus. These geniculate cells have been traced back to their retinal ganglion cell inputs, which have their dendritic arbors in the on-lamina of the inner plexiform layer, quite different than the bistratified arbors of S on/ L off ganglion cells. This produces a curious difference in the blue/yellow channels of color vision in primates. In small animals, ground squirrels, guinea pigs and mice, there appears to be a more symmetrical system of S cone on/Lcone off and M cone on/S cone off opponent ganglion cells, both of which send their dendrites to the on-lamina of the inner plexiform layer. This implies that the horizontal cells and/or intervening amacrine cells are involved in their unique opponent organization, perhaps more similar to the rod system. The existence of this retinal L cone on/S cone off channel would preclude the midget cells depicted in Figure 7 from contributing to color vision in animals with divariant color vision..

A more iconoclastic hypothesis recently proposed by the Neitz group considers the bistratified S cone on/L cone off retinal ganglion cell to play no role in color vision proposing that its signals go to the brain stem where they mediate unconscious visual functions together with light sensitive retinal melanopsin ganglion cells. They propose that surround antagonism from the H2 horizontal cell turns L cone off-center midget bipolar cells into S cone-on/L cone-off cells and L cone-on/S cone-off cells. This hypothesis brings S cone signals into the midget ganglion cell system designed for high spatial resolution. Evidence that the bistratified S cone on-cell does affect conscious vision comes from an observation of W.S. Stiles (1949) that his S cone mechanism, isolated psychophysically, affects conscious experience because it exhibits a curious behavior its sensitivity declines when a long wavelength adapting light, which has no effect on S cones, is turned off. The bistratified S-cone ganglion cell exhibits this same behavior implying that its signals do reach visual cortex and consciousness. It is also possible, however, that the unusual midget cells proposed by the Neitz team also exhibit this phenomenon, which has been labeled “transient tritanopia”.

Regardless of which of these three hypotheses is correct, there must be a way for the visual cortex to compare L with S cone activity in the same area of visual space in order to produce color vision. In Figure 8 we use two of the ideas that are demonstrated in Figure 7 to illustrate how striate cortex uses these retino-geniculate inputs to construct cells that are more responsive to color.

Figure 9 shows a logical way to extract the signals from the cone mosaic into parallel channels mediating achromatic contrast with high spatial resolution and chromatic contrast with lower spatial resolution from the same mosaic.

The perception of color mixes both achromatic with chromatic signals to create the combined experience of color. In this arrangement the S cone system only plays a role in chromatic vision while the L cone system contributes to both chromatic and achromatic vision. The repertoire of colors that are produced and their relationship to the activity of the two cone mechanisms involved are shown in Figure 10 these colors are shown at borders of minimum achromatic contrast where chromatic contrast becomes most important.

7. Color Constancy and Double Opponency

The schemes of the previous figures neglect a problem called “color constancy”. We see colors as unchanged even when there are large changes in the spectral properties of an illuminant. The colors in a scene illuminated by fluorescent lights, which generate much short wavelength energy or by tungsten filament lights, which generate much long wavelength energy are not significantly altered by such changes. In other words we should see things “bluer” in fluorescent light and redder in tungsten light but we don’t. This constancy of colors despite changes in illumination intrigued Edwin Land, the founder of the Polaroid Corporation, who spent years investigating this phenomenon and demonstrating significant global aspects of color vision. Local objects can reflect identical spectral components from their surfaces but will appear of different color because of the influence of the entire scene. He proposed a model in which the signals from each cone mechanism are normalized over the entire visual scene before being compared with each other locally to generate the perception of the color of an object scene. Figure 11 illustrates why this idea is reasonable. Here two lights beams, one white, the other yellow illuminate a screen.

An arrow blocks a portion of the yellow beam and this produces a shadow that appears “blue”. If one determines the light energy being absorbed by the L and S cones from the screen, it is apparent that there is more light affecting the L than the S cones coming from the shadow, which would suggest that the shadow should look yellow and not blue. But if one normalizes the light coming from the entire screen to 100%, then the relative effect on the S cones becomes greater than that on the L cones and the shadow appears “blue”. This supports the idea that responses from each cone mechanism across the entire visual scene are first normalized before being compared with each other locally to produce the perception of color from specific objects in the scene. Such normalization could occur if S cone and L cone transmitting neurons were to inhibit each other following the rules outlined in Figure 12. This could reduce the responses to global illumination that affected one cone mechanism more strongly than the other and support color constancy. If, in addition, the excitation of these cone specific channels were organized so that each local area of color space would be enhanced by neighboring spectral contrast between cone mechanisms it would lead to simultaneous color contrast.

By the rules of Figure 12, a unit area of color space A would appear “blue” if there were short wavelengths and no long wavelengths in area A the converse would lead to “yellow”. This blueness or yellowness would be enhanced if the surrounding areas of color space were activated by the opposite color, i.e. simultaneous color contrast. Such cells have been called “double opponent” cells because they have one form of antagonism between cone mechanisms mediating a local area of color space and an opposite arrangement in neighboring areas of color space. Charles Michaels found such cells in striate cortex of non-human primates following up on observations of Nigel Daw in goldfish retina. Double opponency tends to separate spectral from energy contrast.

8. Trivariant Color Vision

In primates, high resolution vision and trivariant color vision evolved to enhance survival. A fovea formed to facilitate high resolution achromatic vision and a third opsin evolved from the original mammalian L cone opsin to create a new dimension of color in higher primates (Jacobs, 2008). The gene for the L cone opsin duplicated itself and one of the paired genes developed polymorphisms to absorb further into the long wavelength region of the spectrum. The original long wave sensitive L cone now became an M cone being the partner of an even longer wavelength sensitive L cone. Figure 13 shows that a trivariant system can detect a larger variety of spectral reflectances (Mollon, 1989). In Figure 13 above right, one sees that a divariant color vision system can detect spectral contrasts that reflect more at one end than at the other end of the spectrum these reflectances tend to tilt the solar spectrum.

If, however, a surface reflects less from both sides of the visible spectrum, i.e. bending the solar spectrum, it might be invisible to a divariant system because both cone types could be absorbing the same amount of light from the object and its background (Figure 13 lower left). A trivariant system detects this object because it is impossible for all three cones to be absorbing the same amount of light from the object and its background (Figure 13 lower right). More complex reflecting surfaces might confuse even a trivariant system but they are probably very rare in the natural world. This change split the yellow region of the original spectrum and created two new chromatic percepts, red and green (Figure 14). It is noteworthy that a rise in the beta-band absorption of this new L cone pigment also provided a long wave influence at the short wavelength region of the spectrum.

Figure 15 shows how a trivariant system facilitates the distinguishing of red and green objects which would remain indistinguishable for a divariant observer.

In the fovea, the same midget cells which had been contacting only single L cones for divariant color vision now contacted either an L or an M cone (Figure 16), providing the brain with separate channels for these two different cones and another opportunity for chromatic contrast.

In Figure 16 it is assumed that the bistratified S cone on cell conveys S cone signals from the retina and the midget cell system conveys the signals of L or M cones for color vision (Kolb, 1991). The midget cell system is currently assumed to play a role as a “double duty” detector contributing to both high spatial resolution achromatic vision and lower spatial resolution chromatic vision.

Figure 17 (upper) shows the responses of a midget ganglion cell receiving an excitatory L cone input. This cell is continuously excited by the red adapting field under these conditions the cell does not respond to the small red spot. A blue adapting field stops this continuous discharge and under these circumstances the cell responds to the small red spot. Figure 17 (lower) shows the responses of an L cone on midget ganglion cell (large amplitude) and a M cone on midget-like ganglion cell (small amplitude) responding to a small red spot in the presence of the same adapting fields. The red adapting field continuously excites the L cone cell and inhibits the M cone cell under these conditions the small red spot inhibits the L cone cell, an inhibition mediated by M cones.

Figure 18 shows an S cone on-cell responding to spectrally narrow stimuli.

The cell is excited by the blue spot on the yellow background and is profoundly excited by the turning off of the yellow adapting light. The cell is not responsive to the red spot in the presence of the yellow adapting light but is inhibited by the red spot in the absence of the adapting light.

Figure 19 illustrates how these midget cells could participate in both chromatic and achromatic vision. The same concentrically organized retinal and geniculate cells feed two different circuits in striate cortex. One achromatic with a large number of orientation selective neurons capable of high spatial resolution and the other chromatic built up from the same center/surround organized midget-like cells. How center/surround cells can build up a circuit which is more co-extensively organized more suitable for chromatic processing remains enigmatic.

David Hubel (1987) and the late Bob Rodieck (1988) have suggested that the midget system may not be involved in mediating chromatic vision but another much less common ganglion cell type with a receptive field in which L and M cones driving these cells are co-extensively rather than concentrically organized in the receptive field of these cells. It has been difficult to obtain any evidence for such a cell.

The evolution of trivariant color vision in higher primates increased the repertoire of colors we perceive and the power of spectral contrast to detect objects. The original blue/yellow form of color vision was now accompanied by a parallel system of red/green colors occurring in the yellow region of the spectrum where brightness is maximal (Mullen and Kingdom, 2002). Where the dynamic range of achromatic contrasts are maximal is where chromatic contrast can contribute most to detecting borders when achromatic contrast is minimal. Figure 20 shows how the activity of this trivariant cone system contributes to the variety of major colors we see. The borders of effective energy contrast for each of these colors are minimal, which is where color contrast is especially important.

9. The Cone Mosaic of Trivariant Color Vision

The division of a single L cone type into two spectrally different L and M cones produced a mosaic of L, M and S cones from which each ganglion cell picks out the input of one or the other of these cone mechanisms to transmit to the brain. Adaptive optics has revealed significant differences in the number of L versus M cones in normal subjects (Hofner et al., 2005) (Figure 21).

Some subjects have 15 times as many L than M cones and others have twice as many M than L cones but all have normal color vision. This reinforces the idea that a unit area of color space involves comparisons of large groups of cones. What remains problematic is how midget-like ganglion cells in the parafovea select only L or only M cones to form synaptic contact within this variegated cone field.

10. A parallel system of achromatic ganglion cells

In addition to midget and midget-like ganglion cells described there is another parallel system of larger ganglion cells with less representation in the fovea than the midget cell system (Gouras, 1969). These ganglion cells also have: on- and off-varieties mediated by a separate set of cone bipolar cells and are known to be the “parasol” ganglion cells found by the Golgi method of silver impregnation. They seem to receive their inputs from only L and M cones through a different set of bipolar cells than those serving the midget cell system. Figure 22 shows two retinal ganglion cells in rhesus monkey retina.

The cell with smaller amplitude is a phasic on-cell that responds at on to all three wavelengths with a shorter latency than the tonic S cone on cell which is excited only by the deep blue light (419 nm) and inhibited by the longer wavelengths (610 and 509 nm). It is thought that a separate relatively wide field bipolar cell feeds these phasic cells. These cells receive similar inputs from both L and M cones and therefore cannot play a role in “red/green” color vision and probably no role in blue/yellow color vision. Figure 23 is evidence that they receive no input from S cones. The threshold spectral function obtained on a strong yellow adapting field shows no evidence of an S cone input. The S cone on-cell has a threshold spectral function that shows it lowest threshold to blue light. The phasic cells have their lowest threshold resembling that of the photopic luminosity function.

The retinal latencies at of these two types of cells show a significant difference between the tonic and phasic cells indicating a faster transmission of the latter through the inner nuclear layer by the phasic system. In addition the larger size of the axons of the phasic system allows this system to transmit its signals to visual cortex much quicker than the tonic system. The role of the phasic “parasol” ganglion cell system in vision is not entirely clear. It seems to play a role in the detection of motion and to target different areas of visual cortex. Whether it plays any role in color vision is moot.

11. The Lateral Geniculate Nucleus (LGN)

The LGN reorganizes the parallel ganglion cell systems serving local retinal areas into separate layers from where their axons project to specific layers in striate cortex (Martin et al., 1997: Chatterjee and Calloway, 2003, Tailby et al., 2008). Signals from nasal retina of the contralateral eye are separated into different layers from those coming from the temporal area of the ipsilateral eye. This is needed for stereopsis where changes in eye position require comparing signals from different corresponding retinal regions and accordingly different ganglion cells. Binocular interactions begin for the first time in visual cortex. The parallel channels coming from the same retina are also separated into different layers. The phasic achromatic ganglion cell system goes to the magno-cellular layers, one layer for each eye. The midget and midget-like retinal ganglion cells transmitting signals for high resolution achromatic vision and “red-green” color vision are placed in four parvo-cellular layers, two for each eye. The S-cone mediating retinal ganglion cells which are involved in “blue-yellow” color vision go to the konio-cellular layers. These parallel systems of ganglion cells project to specific layers of striate cortex where form, color, movement, direction and stereopsis are processed.

An interesting but unknown role must be played by the numerous centrifugal axons that arise from striate cortical layers and impinge upon the cells in the LGN. Whether this shapes or influences chromatic and achromatic contrasts that are involved in form vision is unknown.

12. Striate Cortex

The LGN transmits its signals to striate cortex, the first visual area (V1) to process visual signals in cerebral cortex. There is a retinotopic order to the projection of these producing a map of the visual field. The map is distorted because the fovea occupies a relatively large area compared to the more peripheral retina. There are columns of cells extending from the upper to the lower layers, one column receiving signals from one eye and a neighboring column receiving signals from the other eye (Figure 24). Within each of these “ocularity” columns there are micro-columns of cells favoring a particular orientation of an extended light stimulus. Together these sets of columns have been called a “hyper-column”. Within each hyper-column there is a local area that receives inputs from chromatic selective cells, i.e. the S cone on and off cells and the L and M cone midget and midget-like cells these areas have been called “blobs”, and are thought to be where color is processed.

There are six distinct layers extending from an upper layer 1 down to a lowermost layer, 6. The projections from the functionally different layers of the LGN target different layers to synapse on cortical cells (Figure 24).

The L and M midget and midget-like cells, projecting from the parvo-cellular layers, synapse in layer 4Cbeta from which their post-synaptic target cells appear to send axons to layer 3 within “blobs”. The S cone system, projecting from the konio-cellular layers, appears to send axons directly to layer 3 within blobs. The fact that the parvo-cellular input, incorporating the L and M cone midget and midget-like cells, makes an intermediary synapse may be due to the double duty role these cells play in handling both achromatic and chromatic vision. The area surrounding blobs process achromatic contrast, undoubtedly receiving their input from the parvo-cellular layers. The magno-cellular layers of the LGN target cells in layer 4Calpha from which postsynaptic cells send axons to layer 4B. Therefore there is segregation of cell systems, some playing a role in chromatic contrast and others playing a role in achromatic contrast and movement. The existence of “double opponent” cells in blobs supports the idea that this zone within each hyper-column is devoted to detecting chromatic contrast for color vision. There are many unknowns in this structure that can only be solved by more single cell physiology combined with anatomical insights into the circuitry. What seems clear is that each unit area of visual space is processed in parallel by achromatic and chromatic mechanisms (Livingston and Hubel, 1987).

13. Pure Spectral Contrast

Double opponent cells tend to eliminate the influence of energy contrast establishing pure spectral contrast as an indicator of the borders of objects. With trivariant color vision this can be formed by comparing L with M cone responses and/or by S cone versus L&M cone responses. Figure 25 illustrates the organization of a “red/green” double opponent cell. The system uses spectral contrast.

Figure 25 shows how concentrically organized double opponent cells can be constructed by having identical (red) neighboring cells inhibit a central cell (red) and have opposite neighboring cells (green) excite the central cell (red) making it most sensitive to a red central object in a green background and unresponsive to the converse. Such a cell responds to spectral rather than energy contrast as an indicator of a border. This behavior is illustrated by a cell detected in visual cortex that responds exclusively to spectral and not to energy contrast (Figure 26). The stimuli with maximum energy contrasts such as a red or green bar on a black background or a black bar on a green or red background produce no response from the cell. But a green bar on a red background where energy contrast is minimal generates a large response. The brain has now separated the two major forms of contrast, spectral versus energy contrast to use them as independent variables to create color.

14. Simple, Complex and Hypercomplex Double Opponent Cells

One of the remarkable discoveries in visual cortex has been that of a unique body of cells that exhibit definable patterns of responses all based on orientation selectivity. One group called “simple” cells show orientation selectivity which can be predicted by the organization of responses produced within their receptive field. Another group, called “complex” cells, which also show orientation selectivity but are less obviously related to the distribution of responses within their receptive field. A third group, “hypercomplex” cells, are orientation selective but are “end-stopped” being inhibited if the oriented stimulus is too long. These three types of cells are thought to contribute to form vision based on energy contrast. Charles Michael (1981) has shown that a similar group of double opponent cells exist in striate cortex of monkeys. Figure 27 shows such a simple cell responding best to strong spectral contrast.

Figure 27 shows that this cell is excited when an oriented red bar moving upward crosses the red selective area of its receptive field and inhibited when it crosses the green selective area it produces an off-response when it leaves this green selective area. A green bar produces excitation when it crosses the green selective area only. A green-red and a red-green bar produce strong responses but the responses occur at different points in the receptive field and are therefore totally out of phase with each other. This cell responds poorly to white bars being spectrally selective. Figure 28 shows a complex cell that responds to spectral and not energy contrasts. It is both orientation and directionally selective and refuses to respond to energy but responds strongly to spectral “red/green” contrast.

Hypercomplex double opponent cells (Figure 29) are also detectable in striate cortex completing the variety that are based on energy contrasts. Therefore the same repertoire of orientation selective cells responsive to energy contrasts in striate cortex can also be found for spectral contrast. It is thought that both groups of cells play a role in form vision. Parallel systems of cells are detecting the forms of objects, one based on energy and the others based on spectral contrast. Because energy contrasts allow a higher spatial resolution, more of visual cortex is dedicated to achromatic than chromatic contrast detection.

Figure 30 shows how these two parallel contrast detecting systems could work in detecting a red object. Both simple cells of different orientation selectivity can detect the border of this cross, one set based on energy contrast and with a higher spatial resolution and the other set based on spectral contrast and with a lower spatial resolution. Sharp corners in the object would be detected by end-stopped hypercomplex cells. This creates two different views of the object which may be fused into one colored object just as two similar stereoscopic views are fused into an object in depth. Together they create shapes in full color (Figure 30).

A similar system for using spectral contrast created by double opponent S cone versus L cones or L+M cones must create another spectral view of objects by border contrasts (Figures 31 and 32). In Figure 31 groups of double opponent simple cells detect a circular blue form on a yellow background. In Figure 32, similar simple cells detect a border of contrast between yellow and white, allowing the detection of an object of minimum energy contrast.

This highlights an important role of the S cone mechanism in vision to detect border contrasts of white versus yellow which are difficult for the achromatic system which is mediated only by L and M cones. This scheme implies that for color vision three different views of the same object are formed and then fused to provide colors that represent mixtures of the red/green and the blue/yellow spectral contrasts systems. If there is no activity in the spectral contrast detectors then only the energy (achromatic) system is active and the object is pure white, gray or black. There are some different views on these ideas. Single opponent cells are thought by Johnson et al, (2008) to detect surface color while double opponent cells detect borders. Others have suggested that neurons involved in the analysis of color should retain their chromatic properties in the face of changes in other components of stimulation, orientation, size and contrast and this may not be the case for color selective neurons found in striate cortex (Solomon & Lennie, 2007).

15. The Stabilized Retinal Image

If an image is made stationary on the retina it will disappear in seconds. Figure 33 illustrates what is perceived by a subject presented with a stabilize image of a red/green boundary based on the research of the Russian physiologist Yarbus (1967) and others (Ditchburn, 1973). Initially the subject sees the red/green scene but in several seconds it fades into a featureless field. If a pale blue light is added to both sides only a blue field is seen without the red/green scene. This also fades within seconds. If the pale blue field is removed the subject sees a faint red/green scene but this fades again. This indicates that we only see in transients determined by change in the retinal image.

If the retinal image does not change in space and/or time, it will disappear. Constant small micro-movements of the eye continue to produce transient changes at the borders of objects that are able to maintain an undiminished view of the external world. This supports the idea that objects are seen because of their borders, as depicted by the orientation selective border contrast cells illustrated previously, which must be continuously moved across the retina to maintain their visibility. Figure 34 illustrates how the responses of on and off cells contribute to maintaining strong responses to movements of borders of contrast. When a long wavelength (red) surface enters the receptive fields of on and off cells, the on-cell (L off) responds more strongly than the off-cell if there is an increment in energy contrast. But the off-cell (M off) will respond more strongly if there is a decrement in energy contrast. These two systems are showing their ability to respond to increments and decrements of energy contrast but also manage to favor long over other shorter wavelengths, which is a key for their role in color vision. M on and L off -cells would respond weakly to such a moving border.

16. Color Vision Beyond Striate Cortex

An intriguing but poorly understood aspect of visual cortex is the remapping of visual space into multiple adjacent visual areas where a retinotopic order is maintained. Figure 33 shows how V1 projects to V2, a second visual area juxtaposed to V1 where there continues to be a segregation of functionally different cells. The “blob” areas of V1 appear to project to “slabs” of cells in V2, which project to “globs” of cells in V4.

Within these three interconnected areas spectral contrast is processed, presumably to determine its contribution to color, i.e. hue. Adjacent areas processing energy (achromatic) contrasts contribute to determining the saturation and brightness in colors. The achromatic areas are larger because of the higher spatial resolution of energy contrasts. The receptive field sizes of cells in these higher visual areas tend to be larger than they are in V1 suggesting that, several blobs in V1 project to a single slab in V2 and several slabs in V2 project to a single glob in V4 (Conway et al., 2007 2006 2002). The functional roles of these higher areas could be to recognize objects better as they come closer to the observer when they cast a larger retinal image. It is known that the perception of color increases with the size of an object over sizes as large as 20 degrees of visual angle. Under ideal conditions i.e. a well lit surface, we can perceive at least a million different colors. Figure 36 shows how more basic hues determined by single cells in blobs could be used to build higher order colors that combine the inputs of the more basic hue determining cells. As one develops more experience with colors even higher order cells could be formed allowing the distinguishing of many different shades of color.

There is a question as to whether a single cell is committed to detecting a particular color or whether a group of cells are involved. This might depend on the practice one has in working with colors such as artists who may form single cells that are sensitive to a particular color and capable of contributing to even higher order colors while those who are less involved have fewer such cells and depend on less fastidious interactions.

There has been a controversy about whether color processing is exclusively segregated to V4 while achromatic processing occurs in another complete area of visual space. Evidence implying this type of organization involves rare individuals who have lost all color vision but can see objects with high spatial resolution in black and white. This hypothesis was championed by Zemir Seki but the existence of “globs” related to spectral contrast in V4 surrounded by larger areas of achromatic contrast detecting cells tends to weaken this idea. Because “globs” are relatively large it was perhaps possible for an investigator making a limited number of penetrations to conclude that all of V4 is devoted to chromatic vision. Nevertheless there must be an explanation of why in these very rare patients color can be lost without a loss in spatial resolution which depends entirely on achromatic contrast. The enigma may be explicable by damage to higher areas of the brain where interpretation and language are involved linking vision to external communication.

17. Redness at short wavelengths

It is well known that short wavelengths can produce a reddish sensation that makes blue appear violet. This affect appears to arise from the beta-band of L cone opsin which makes this opsin more sensitive to light than M cone opsin at this region of the spectrum (see Figure 14). This effect is best observed in midget-like cone opponent retinal ganglion cells which are excited by long wavelengths and inhibited at wavelengths at the middle of the spectrum. In many of these cells excitation appears with short wavelength stimulation implying that the L cone response is overpowering the M cone antagonism. This is supported by the fact the L cone response is strengthened by a blue adapting light, which selectively depresses the responses of M cones. Had this short wavelength excitatory input been due to S cones, it would have been weakened by the blue adapting light.

18. Hering’s Theory of Color Vision

An insightful theory was proposed by Ewald Hering in Prague in 1874 which was resisted by many leading scientists in the field, such as Herman von Helmholtz. Hering noted that the sensations of blue and yellow opposed each other so that when seen simultaneously, the original colors were lost and a new sensation of a totally different color “white” was seen. Similarly the colors red and green exhibited a similar opposing relationship. When mixed they lose all traces of the original colors to create a new sensation “yellow”. Hering suggested that there were two pairs of opponent processes underlying human color vision. This contrasted with the fact that there are only three variables underlying human color vision. This three dimensional system was suggested by Thomas Young (1802) and proven by James Clerk Maxwell (1872) and strongly promulgated by Helmholtz (1852). The Hering theory suggested that there are two pairs of basic colors that have a unique opponent relationship for each pair. He was correct in this analysis although his idea that the opponency between colors was occurring in the photoreceptors was incorrect.

At Hering’s time the understanding of neuro-physiology was poor and the concepts of excitation and inhibition of neurons in unique and complex ways was unknown. The first suggestion that antagonism between colors could be due to excitation and inhibition between nerve cells, that I have found, was in a paper by the Swedish physiologist Gustaf F. Gothlin (1944). He had followed Sherington’s work on the spinal cord where the idea of excitation and inhibition between nerve cells first evolved. Figure 37 shows how prophetic Gothlin’s ideas were. Here he creates an antagonistic balance between blue and yellow processes as a first stage and then hangs a scale on the yellow side of the blue/yellow balance to show an antagonistic relationship between red and green. He predicts that these antagonistic interactions were due to excitation and inhibition between neurons. It is interesting that the pioneer of exploring nerve cell responses to spectral stimuli in the retina, the Nobel Prize winner, Ragnar Granit, also a Swedish physiologist, made no reference of Gothlin’s concept in any of his writings on retinal responses to color.

Granit (1947) proposed a dominator/modular theory of color vision in which there were some retinal neurons which responded to only a narrow part of the visible spectrum, “modulators”, and another group that responded to a broad band of the spectrum, “dominators”. He thought the modulators were more involved in color vision but never mentioned the concept of excitation and inhibition between colors or cone mechanisms as Gothlin did. The first American studies of single visual neuron responses to monochromatic stimuli were obtained in the monkey’s lateral geniculate nucleus by Russel DeValois and associates in 1958. They emphasized the presence of narrow-band responding cells resembling the modulators of Granit but they also noted some cells that gave on responses to red and off responses to green light but again Hering’s concept of opponent colors was not mentioned. The experiments that brought Hering’s opponent color theory to the forefront were done in fish retina by another Swede, Gunter Svaetichin (1956) using glass instead of metal electrodes, enabling him to record intra-cellularly from horizontal cells. This allowed him to see both depolarizing and hyperpolarizing responses, the latter not obvious with extra-cellular recordings, where one sees only a silence or perhaps an off response from inhibition (hyperpolarizing) which could easily be overlooked. Svaetichin’s results gave rise to a resurgence in Hering’s opponent color theory (Hurvich and Jamison, 1957 Hurvich, 1981). This quickly led to the idea that there were three retinal channels representing Hering’s color opponent channels, red opposing green, yellow opposing blue and white opposing black. This idea, however, was short lived. First of all it became apparent that the red/green opponent cells had concentrically organized receptive fields with the color opponent inputs coming from adjacent retinal regions. One would expect the color opponent signal should involve the same area of visual space and not neighboring areas. In addition the red-green opponent ganglion cells seemed to be the midget cell system which was the logical mediator of high spatial resolution. This created a dilemma in having a red/green color opponent channel carrying achromatic information for high spatial resolution. In addition the S cone channel was opposed by yellow light but not strongly enough to stop responses to white. The S cone on ganglion cells appeared to be transmitting a signal that said that the S cones were absorbing light but not that this light was “blue” it could be white, grey as well. The so called white/black channel was receiving no input from S cones which is not what is expected for a channel signaling white. The contribution of activation of the S cone retinal ganglion cells is illustrate by the after images produced in Figure 38.

Therefore the link between Hering’s opponent color theory and retinal ganglion cell signals was scrapped. Because the neurons in the lateral geniculate nucleus responded similarly to what was found in the retina, attention went to striate cortex to explain opponent color theory by physiology. The initial ideas of channels signaling color at the retinal level were wrong the opponent signals detectable in the retina were cone opponent and not color opponent. Color vision turned out to be more sophisticated, more arcane but more realistic.

19. The Future

Certainly the future will involve anatomical definition of how the projections from the lateral geniculate synapse in striate cortex and begin to define the neurocircuitry of this hexalaminar cortical structure. The dendritic input and axonal output of each functionally unique cell must be defined as has been done to a large degree in the retina. This strategy must be repeated in higher areas of visual cortex, perhaps the most intriguing phase of this crusade. New optical techniques that combine both anatomical identification and physiological responses of the same cells could expedite this enormous task. But this is the only way to understand how the most complex machine in our universe, the cerebral cortex works. For this color vision is an ideal starting point because it has carried us further into this organ than any other neural process.

20. References

Chatterjee S, Callaway EM. Parallel colour opponent pathways to primary visual cortex. Nature. 2003426:668–671. [PubMed]

Conway BR, Hubel DH, Livingstone MS. Color contrast in macaque V1. Cereb Cortex. 200212:915–925. [PubMed]

Conway BR, Livingstone MS. Spatial and temporal properties of cone signals in alert macaque primaryvisual cortex. J Neurosci. 200626:10826–10846. [PubMed]

Conway BR, Moeller S, Tsao DV. Specialized color modules in Macaque extrastriate cortex. Neuron. 200756:560–573. [PubMed]

Dacey DM, Lee BB. The “blue-on” opponent pathway in primate retina originates from a distinct bistratified cell type. Nature. 1994367:731–735. [PubMed]

Daw NW. The psychology and physiology of colour vision. Trends in Neurosci. 19847:330–335.

De Valois RL, Smith CJ, Kitai ST, Karoly AJ. Responses of single cells in monkey lateral geniculate nucleus to monochromatic light. Science. 1958127:238–239.[PubMed]

Ditchburn R. Eye Movements and Visual Perception. Claredon Press 1973.

Ekesten B, Gouras P. Cone and rod inputs to murine retinal ganglion cells: Evidence of cone opsin specific channels. Vis Neurosci. 200522:893–903. [PubMed]

Gothlin GF. Interaction of excitatory and inhibitory processes in the synthesis of the sensation of colour and of white. Upsala Lek refurenings Furhandlingar. 1944 49: 433-446.

Gouras P. Antidromic stimulataion of orthodromically identified ganglion cells in monkey retina. J Physiol. 1969204:407–419. [PubMed]

Granit R. Sensory Mechanisms of the Retina. London: Oxford University Press 1947.

Haverkamp S, Wassle H, Duebel J, Kuner T, Augustine GJ, Feng G, Euler T. The primordial, blue-cone color system of the mouse retina. J Neurosci. 2005 25:5438 -5445.22:3831-3843.

Helmholtz HLF von. Handbuch der Physiologischen Optik, 2nd edition. Hamburg: Voss. (3rd edition, Hamburg:Voss, 1909) 1852.

Hering E. Outlines of a theory of the light sense (translated by Leo M Hurvich and Dorothea Jameson). Cambridge (MA): Harvard University Press 1964.

Hofer H, Carroll J, Neitz J, Neitz M, Williams DR. Organization of the human trichromatic cone mosaic. J Neurosci. 200525:9669–9679. [PubMed]

Hubel DH. Eye, Brain and Vision. NY: Scientific American Library 1988.

Hurvich L. Color Vision. Sunderland (MA): Sinauer Associates 1981.44.Hurvich LM, Jameson D. An opponent-process theory of color vision. Psychol Rev. 195764:384–404. [PubMed]

Jacobs GH. Primate color vision: a comparative perspective. Vis Neurosci. 200825(5-6):619–33. [PubMed]

Johnson EN, Hawken MJ, Shapley R. The orientation selectivity of color-responsive neurons in macaque V1. J Neurosci. 200828:8096–8106. [PubMed]

Kolb H. Anatomical pathways for color vision in the human retina. Visual Neurosci. 19917:61–74.

Land EH. An alternative technique for the computation of the designators in the Retinex theory of color vision. Proc Natl Acad Sci USA. 198683:3078–3085.[PubMed]

Lee BB. Neural models and physiological reality. Vis Neurosci. 200825:231–41. [PubMed]

Livingstone MS, Hubel DH. Psychophysical evidence for separate channels for the perception of form, color, movement and depth. J Neurosci. 198711:3416–68.

Martin PR., White AJR, Goodchild AK, Wilder H D, Sefton A E. Evidence that blue-on cells are partof the third geniculocortical pathway in primates. Eur. J. Neurosci. 19979:1536–1541. [PubMed]

Maxwell JC. On colour vision. Proc R Inst. Great Britain. 1872 6:260-271.

Michael CR. Columnar organization of color cells in monkey’s striate cortex. J Neurophysiol. 198146:587–604. [PubMed]

Mollon JD. “Tho’ she kneel’d in that place where they grew. . .” The uses and origins of primate colorvision. J Exp Biol. 1989146:21–38. [PubMed]

Mullen KT, Kingdom FA. Differential distributions of red-green and blue-yellow cone opponency across the visual field. Vis Neurosci. 200219:109–118. [PubMed]

Neitz J, Neitz M. Colour vision: the wonder of hue. Curr Biol. 200826(18):R700–702.

Nelson R, Famiglietti EV, Kolb H. Intracellular staining reveals different levels of stratification for on-center and off-center ganglion cells in the cat retina. J Neurophyiol. 197841:427–483.

Packer O, Verweij J, Schnapf JL, Dacey DM. Primate S cones have blue-yellow opponent receptive fields. Invest Ophthalmol Vis Sci. 200748:2849.

Polyak, SL. The Retina. Chicago: Univ. Chicago Press 1941.

Rodieck RW. The primate retina. In: Comparative Primate Biology Series, Volume 4: Neurosciences (eds. Steklis H and Erwin J). New York: Alan R. Liss. 1988 pp. 203-278.

Solomon SG, Lennie P. The Machinery of Colour Vision. Nature Reviews Neurosci. 20078:275–286.31.Yarbus AL. Eye Movements and Vision. New York: Plenum Press 1967.

Stiles WS. Increment thresholds and the mechanisms of colour vision. Dcoumenta Ophth. 19493:138–165.

Tailby C, Szmajda BA, Buzas P, Lee BB, Martin PR. Transmission of blue (S) cone signals through the primate lateral geniculate nucleus. J Physiol.2008586:5947–67. [PubMed]

Wright WD. Researches on Normal and Defective Colour Vision. London: Henry Kimpton 1946.

Yin L, Smith RG, Sterling P, Brainard DH. Physiology and morphology of Color-Opponent ganglion cells in a retina expressing a dual gradient of S and M opsins. J Neurosci. 200929:2706–2724. [PubMed]

Young T. On theory of light and colours. Phil Trans R Soc. 180292:12–48.

Seki Z. A vision of the brain. Oxford: Blackwell Scientific Publications 1993.


A New Study About Color Tries to Decode ‘The Brain’s Pantone’

To revist this article, visit My Profile, then View saved stories.

Photograph: Ilona Nagy/Getty Images

To revist this article, visit My Profile, then View saved stories.

Bevil Conway, an artist and neuroscience researcher at the National Institutes of Health, is crazy about color. He particularly loves watercolors made by the company Holbein. “They have really nice purples that you can’t get in other paints,” he says. If Conway is after a specific shade—perhaps the dark, almost-brown color the company has labeled “Mars Violet” or the more merlot-tinted “Quinacridone Violet”—he might scroll through a Holbein chart that organizes the colors by similarity. Anyone who has considered painting a wall is familiar with these arrays: lines of color that transition from bright yellows into greens, blues, purples, and browns.

But if Conway decides to shop around at another paint company like Pantone, that chart, also known as a “color space,” will be organized differently. And if he chooses to consult the Commission Internationale de l’Éclarage, an organization that researches and standardizes measurements of light and color, he will find yet another unique map. Conway is baffled by the choices. “Why are there so many different color spaces?” he asks. “If this is really reflective of something fundamental about how we see and perceive, then shouldn’t there be one color space?”

How humans perceive color, and how all those shades are related, is a question scientists and philosophers have been attempting to answer for millennia. The ancient Greeks, who famously had no word for the color blue, argued over whether colors were composed of red, black, white, and light (that was Plato’s theory), or whether color was celestial light sent down from the heavens by the gods and that each color was a mixture of white and black or lightness and darkness (that was Aristotle’s). Isaac Newton’s experiments with prisms identified the components of the rainbow and led him to theorize that the three primary colors, from which all other colors are made, are red, yellow, and blue.

Today, our scientific understanding of color perception is rooted in biology. Each color represents a specific part of the electromagnetic spectrum, though humans can only see the slice of this spectrum known as “visible light.” Of the wavelengths visible to humans, red ones are longer, while blues and violets are shorter. Photons of light stimulate photoreceptors in the eye, which transform that information into electrical signals that are sent to the retina, which processes those signals and sends them along to the brain’s visual cortex. But the mechanics of how the eye and nervous system interact with those light waves, and how a person subjectively perceives color, are two very different things.

“One way to think about neuroscience is that it’s a study of signal transformations,” writes Soumya Chatterjee, a senior scientist at the Allen Institute for Brain Science who studies the neurology of color perception, in an email to WIRED. He says that once the photoreceptors in the retina have passed information to the visual cortex, the information continues to be transformed—and scientists don’t yet understand how those series of transformations give rise to perception, or the experience an individual person has of color.

Some aspects of color can already be measured precisely. Scientists can calculate the wavelength of the light and the luminance, or brightness, of a color. But once you bring human perception into the mix, things get a little more complicated. People perceive color by factoring in a number of other variables, like the quality of the light or the other tones bordering the color. Sometimes that means the brain will perceive the same object as two completely different colors that happened with the famous dress, which in some lights looked white and gold and in others looked blue and black.



Comments:

  1. Arashilrajas

    Excuse, that I interfere, but, in my opinion, there is other way of the decision of a question.

  2. Bainbridge

    Great message, very interesting for me :)

  3. Yozshuzragore

    And that we would do without your remarkable phrase



Write a message