the art + science of seeing
volume 2 issue 3 volume 2.3 autumn 2009 Color 1
volume 2 issue 3
the art + science of seeing
Glimpse is an interdisciplinary journal that examines the functions, processes, and effects of vision and vision’s implications for being, knowing and constructing our world(s). Each theme-focused journal issue features articles, visual spreads, interviews and reviews spanning the physical sciences, social sciences, arts and humanities.
Gl mpse
the Color issue
TM
TM
Contents
Glimpse  
www.glimpsejournal.com
T h e Co lor i ssue
4
10
Fluorescence in the garden
40
Christie Marie Bielmeier
14
Ordering Colors A multifaceted problem
Carolyn Arcabascio
46
Rolf G. Kuehni
24
29
retrospect Processes for making the best and finest sort of prussian blue with quick-lime; Concerning the secret of a red gum...and Concerning the source of an illusion... Color Matters
Image courtesy of Jack Shainman Gallery, New York.
seeing red on mars Adaptation and the influence of the environment on color appearance Michael A. Webster
54
waves of color An ecological valence theory of human color preference Karen B. Schloss & Stephen E. Palmer
62
Odili Donald Odita
Cover Images Odili Donald Odita, Fusion, 2006, acrylic on canvas, 96 x 120 inches. OD06.010.
interview with evolutionary biologist Hopi hoekstra
playing (with) color Fred Collopy
68
relatively speaking The relationship between language and thought in the color domain Debi Roberson & J. Richard Hanley
volume 2.3 autumn 2009 Color 5
color-Struck Quilting and colorism in the African-American community Lauren Cross
82
Human potential for tetrachromacy Kimberly A. Jameson
92
singed bedroom, weekend afternoon Arto Vaun
93
watercolor science Transparent watercolor through the eyes of an aerospace engineer Christie Marie Bielmeier
100
(Re)views Blue & The Wizard of Oz Ivy Moylan
online
78
INTERVIEW HP Color Scientist Nathan Maroney Lauren Cross
Contributors
Fred Collopy
Kimberly A. jameson
Dr. Collopy is Professor and Chair of Information Systems and Professor of Cognitive Science at Case Western Reserve University. He received his PhD from the Wharton School of the University of Pennsylvania. He designed the first version of Imager for the Apple II computer in 1977. He has played Imager with the Cleveland jazz ensemble Kassaba, San Francisco experimental composer Henry Warwick, Miami DJ Dino Filipe and with his own Rhythmic Light Ensemble. He has been a visiting scientist in the computer music center at IBM’s Watson Research Lab.
Dr. Jameson is a cognitive scientist conducting research at the Institute for Mathematical Behavioral Sciences, at the University of California, Irvine. Color figures prominently in her empirical and theoretical work, which includes research on the mathematical modeling of color category evolution in societies of communicating robots; individual variation and universals in human color cognition and perception; the genetic underpinnings of color perception, and comparative investigations of the ways the worlds’ cultures name and conceptualize color in the environment.
Rick Hanley
Rolf G. Kuehni
Dr. Hanley has been Professor of Neuropsychology at the University of Essex, UK since 1998. In addition to his work with Dr. Debi Roberson on color, he has a wide range of research interests. He has published several papers that have examined the influence of writing systems on learning to read. They include studies of Chinese, English, Greek, Spanish and Welsh readers. Other current research projects include the way in which we retrieve information about familiar people, and the effect of brain injury on speech production.
Rolf Kuehni is a former chemical industry executive and currently an adjunct professor in color science at North Carolina State University. He is the author of several books on color, in particular, together with A. Schwarz, Color Ordered, A survey of color order systems from antiquity to the present (Oxford University Press, 2008), and of many peer-reviewed articles. Originally from Switzerland, he graduated as a textile chemist from Fachhochschule Niederrhein in Krefeld, Germany.
Hopi Elisabeth Hoekstra
Odili Donald Odita
Dr. Hoekstra is currently the John L. Loeb Associate Professor of Biology in the Department of Organismic and Evolutionary Biology and the Curator of Mammals at the Museum of Comparative Zoology at Harvard University. She is broadly interested in the genetic basis of adaptation and speciation in vertebrates. Her research has primarily used natural populations of rodents to understand the ultimate and proximate causes of evolutionary change.
Odita was born in Enugu, Nigeria and lives and works in Philadelphia and New York. Odita is currently an Associate Professor of Painting at Tyler School of Art, Temple University in Philadelphia. Odita has participated in numerous oneperson and group exhibitions including the Studio Museum in Harlem; Yerba Buena Center for the Arts, San Francisco and the 52nd Venice Biennale International Art Exhibition. Odita received a Louis Comfort Tiffany Foundation Grant in 2007, and a Joan Mitchell Foundation Grant in 2001.
Debi Roberson
Michael A. Webster
Dr. Roberson obtained her PhD from the University of London. Her PhD fieldwork compared color categorization with similarity judgments, memory and learned associations for colors in a hunter-gatherer tribe in Papua New Guinea. Recent research has examined Categorical Perception in adults under verbal interference, or when colored stimuli are presented to one or other visual field (in collaboration with Rick Hanley and Hyensou Pak). Current projects include the development of children’s categorization abilities in populations with typical and atypical language development.
Dr. Webster is a Perceptual Psychologist at the University of Nevada, Reno where he is a Foundation Professor of Psychology. He received his PhD from the University of California, Berkeley in 1988 and was a postdoctoral fellow at the University of Cambridge before moving to Nevada in 1994. In addition to the US and UK, he has also lived in Egypt and travels frequently to India, where he and his wife Shernaaz have conducted studies of the role of culture and the environment on color perception.
Karen B. Schloss Schloss is a graduate student in the Department of Psychology at the University of California, Berkeley. She received her BA from Barnard College in 2005 with a major in Psychology and a minor in Architecture. Her research concerns visual perception, with a concentration on color aesthetics. She has also done work on visual illusions and figure-ground organization.
Color
Arto Vaun is a PhD candidate in Creative Writing at the University of Glasgow. His first book of poems, Capillarity, was published in 2009 (Carcanet Press). One of Vaun’s poems in Capillarity received High Commendation by the Forward Prize as one of the best poems published in the UK in 2009. He has taught at the University of MassachusettsBoston, Glasgow University, and Mass Bay Community College. He is also a songwriter, performing as The Kent 100s.
autumn 2009
Arto VAUN
Dr. Palmer is a Professor of Psychology and Cognitive Science at the University of California, Berkeley, where he has been on the faculty since 1974. His primary area of research is visual perception, where he has made seminal contributions to Gestalt grouping principles, figure-ground perception, and related aspects of perceptual organization. His recent research has focused on visual aesthetics, particularly the study of color and spatial composition.
volume 2.3
Stephen E. Palmer
7
Glimpse the art + science of seeing I am paying by __Check ___VISA ___MasterCard ____________________________________________ Card Number Exp. Date ____________________________________________ Signature Send my subscription to: ____________________________________________ Name (required) ____________________________________________ Email address (required for subscription correspondence)
____________________________________________ Street Address, Apt # (required for print subscription)
SELECT TYPE BELOW ELECTRONIC ____ 1-year/$40 ____ 2-years/$65
INVE ST IN
United States ____1-year/$120 ____ 2-year/$225
INSI
____________________________________________ City, State/Province/County Postal Code
Country
Mail this form with payment to: Glimpse Journal PO Box 382178 Cambridge, MA 02238
Or subscribe online www.glimpsejournal.com/subscribe.html
Canada ____ 1-year/$126 ____ 2-year/$240 International ____ 1-year/$195 ____ 2-year/$380
GH T
The Dawn of the Color Photograph Albert Kahn’s Archives of the Planet
David Okuefuna In 1909 the French banker and philanthropist Albert Kahn launched a monumentally ambitious project: to produce a color photographic record of human life on Earth. Over the next twenty years, he sent a group of photographers to more than fifty countries around the world, amassing more than 72,000 images. Today Kahn’s collection of early color photographs is recognized as one of the world’s most important. “[D]oes the past change when we see it in color? In many instances, the vivid palette brings the images closer to our present moment, making the world—and the distance of history—frighteningly small.” —Nicole Rudick, Bookforum “[A] handsome document full of lush and memorable images. Most of us still picture 1909 exclusively in black and white, so it’s a revelation to peer back 100 years and see such eerily bright hues.” —Dushko Petrovich, The Boston Globe 336 pages. 370 color illus. 9 x 9. Cloth $49.50 978-0-691-13907-4
For sale only in the United States and Canada
800.777.4726 press.princeton.edu
From the Editor
w
Glimpse Team Megan Magenta Hurst
Founder, Managing Editor
Carolyn Cerulean Arcabascio Acquisitions Editor, Interviews
Nicholas Umber Munyan Art Director
Lauren Lemon Yellow Cross Editorial Research, Interviews, Contributing Writer
Christie Chartreuse Marie Bielmeier
This issue is, as with each, a collaboration of many researchers, scientists, scholars, artists, and thinkers assembled by the Glimpse staff (notably, Acquisitions Editor, Carolyn Arcabascio). We thank each of our contributors for sharing their deeply-considered understanding of color. Collectively, they have transformed my own understanding of the ever-present, spectacular phenomenon. If you agree, we encourage you to share the Glimpse Color issue with others, and to “invest in (in)sight” by subscribing to Glimpse, buying a gift subscription for a friend or advertising with us. Watch for Glimpse’s next issue, Cosmos, which will transport this issue’s concerns with electromagnetic waves beyond Earth’s atmosphere, as this International Year of Astronomy draws to a close.
Megan Hurst, Editor editor@glimpsejournal.com
Angie Azure Mah Editorial Intern Ivy Technicolor Moylan Contributor, Film Reviews
Color
Further still, how is our perception of color reinforced or interrupted by language—by our naming of colors? And, since, startlingly, a small percentage of individuals have a gene for tetrachromatic perception (versus the majority of humans who transmit trichromatic perception), will their offspring evolve to see colors that we do not see today? And, what might the long-sought-after symphony of color-correlated-to-sound look and sound like? Can sound amplify our experience of color and vice versa? We conclude with thoughtful reviews of two films in which color is conceptually integral.
Rachel Sepia Sapin Editorial Intern
autumn 2009
Contributing Science Writer, Copy Editor Glimpse vol. 2.3 celebrates what we know about color perception so far, and examines what still eludes us: From our historical efforts at wrangling color into well-understood systems that can be harnessed for our own creativity, to the impressive bio-fluorescence just outside our range of natural perception—but now visible with one engineer’s inventions. We see how the brain’s interpretation of color adapts to our environments, and how our environments may very well adapt to our perception. On a personal level, we learn how skin color influences self- and social-perception (and we can’t help but re-contextualize human “colorism” as petty when considered in a broader biological spectrum). We examine centuries-old recipes for the optimal Prussian blue paint and red inks, and learn one NASA engineer’s scientific method of achieving optimal visual effects in his watercolor paintings.
volume 2.3
For those of us blessed with a full complement of color perception, it’s easy to take color for granted. It is everywhere. Electromagnetic waves of visible spectra bounce around us constantly. To quote Odili Donald Odita, the artist whose striking painting adorns this issue’s front and back covers, “Color matters.” Humans are emotionally moved by color. Color signals to us from nature, and we, in turn, use it to signal to one another. It cloaks us, it accentuates aspects of ourselves, it unites us, it calms us, it excites us, it mesmerizes us. It is as if all organisms and their environments have been tuned to and for each other in a great call and response of color.
9 Arto Plum Vaun Staff Poet + Contributing Poetry Editor
Adjunct + Alumni Christine Madsen Co-Founder, Editor (Europe) EmComm Marketing + Communications Matthew Steven Carlos Editorial Advisor Anthony Owens Photography Jamie Ahlstedt Logo Design
Glimpse PO Box 382178 Cambridge, MA 02238 ISSN 1945-3906 www.glimpsejournal.com Copyright and Acknowledgements Glimpse acknowledges creators’ copyright, and encourages contributors to consider Creative Commons licenses for their works. Many of the images used in this issue are Creative Commons licensed images from Flickr.com members, and others are public domain images courtesy of private collectors. The font used in this issue is Tuffy, a freely available font.
fluorescence in the
Glimpse  
www.glimpsejournal.com
by Christie Marie Bielmeier
10
Garden
volume 2.3 autumn 2009 Color
A
bouquet of red roses tells your sweetheart you are sorry,11 but to a bee it’s a fluorescent green beacon for pollination.
Long thought to attract insects with colors in the humanly-visible
light spectrum, one pioneer of underwater fluorescent photography is discovering that flowers also produce light outside the range of human visual perception. In the darkness of underwater coral reefs, species produce light to help them stand out, and for other purposes. Photographers have captured these striking images from deep in the sea, but now Dr. Charles Mazel, engineer and marine biologist, is collecting images from his own backyard. Using a high-intensity LED flashlight with a custom, proprietary blue-light interference filter and a pair of yellow-tinted glasses, Mazel probes his house for objects that fluoresce. Substances that fluoresce absorb light, or energy, from an external source and, in turn, emit a light of a different color, or wavelength. While Mazel has found that many food items in his kitchen fluoresce—including peanut butter, olive oil and broccoli—most recently he’s been photographing flowers from his garden. Mazel collects his images using a standard camera to photograph a Beetle on White Impatiens flower; (top) photographed in white light; (background) photographed with blue light excitation and matching barrier filter All Photos Courtesy of Dr. Charles Mazel.
subject in natural daylight. Then, returning at night, he uses the same camera with a blue-light filter on the flash and a yellow filter on the camera lens. The resulting images are beautiful—and surprising.
Glimpse
www.glimpsejournal.com
“Our normal visual experience is that everything that reaches our eyes is a combination of all colors—because we see reflected white light,” explains Mazel. Humans are accustoming to viewing objects in a broad range of light, but fluorescence is a narrow range of light, which tends to be pure and saturated.
12
Comparing the white light and blue light photos of flowers, one is drawn to the pistil and stamen area at the center of the petals. Also, the pollen brightly fluoresces. This color placement may be startling since human eyes cannot interpret colors within this range under normal lighting conditions. It’s not important that humans can’t see these details because we don’t pollinate the flowers, but to bees it may be a question of survival. Bees and many insects have ultraviolet vision and photoreceptors that interpret a broad range of wavelengths. The plant’s patterns account for this. To attract bees for pollination, it is advantageous for a plant to have a bull’s-eye with a dark circle around a bright center. Therefore, the plant would grow petals that are ultraviolet-reflective and a pistil or stamen that is UV-absorbent. To create dark areas, the plant produces pigments that absorb ultraviolet light—and as result fluoresce. Current research has proven that plants have these patterns and bees can see them, but which pigments are UV-absorbent is unclear. Mazel hopes to use his life-long experience studying underwater fluorescence to find some of the answers. “I got intrigued by what would happen if you swim around in [the ocean] in the middle of the night with an ultraviolet light,” says Mazel, whose college hobby was underwater photography. Back in the 1980s, the MIT ocean engineering student had little interest in the science of fluorescing plants. Instead, he focused on the aesthetics of the striking images. Mazel began to question whether the green light emitted from one species was the same as another. Mazel says, “I found out that scientists didn’t really know what was going on. They were somewhat aware of the phenomenon of fluorescence in corals, but little scientific data [existed].” This lack of information prompted Mazel to pursue a doctorate in Marine Biology from Boston University. Soon after he began his research, he realized he needed an instrument to scientifically compare fluorescent
colors of two different species—so he engineered one. In 1994, Mazel prototyped the first diveroperated, underwater spectrofluorometer, which was funded by SeaGrant of the National Oceanic and Atmospheric Administration. The handheld instrument measured the excitation created by fluorescing creatures and allowed Mazel to study coral reefs without disturbing the environment. Mazel’s first instruments used ultraviolet (UV) light because, like most people, he associated UV light with black-light posters and mineral displays of fluorescing rocks. Using these instruments, Mazel investigated what wavelengths were most common in fluorescing organisms. After studying spectral data, Mazel was convinced to try filtered blue light instead. From his college lab, Mazel scrounged up some blue light filters and stuck them over a lowpowered light. One night on a dive in the Bahamas, he tried both the high-powered UV light and his cheap blue light. Mazel says, “That weak blue light killed the powerful UV light. More things fluoresced and more brightly. You miss some things—nothing’s perfect, but blue is just way, way better... Fluorescence has more potential to influence color in the ocean because the environment—as you go deeper—tends toward monochromatic.” In water, red, orange and yellow wavelengths become absorbed very quickly. Therefore, fluorescing these colors really makes a species stand out. After graduation, Mazel continued improving his underwater instruments and began working with the Ocean Optics at the Office of Naval Research. While Mazel’s instruments were used for military applications, he was still able to spend time in the ocean. Underwater with his blue-light, Mazel imaged coral reefs that appeared brown in natural light, but exhibited vibrant color under blue light. Also, he found that mantis shrimp, which burrow deep into the ocean floor, showed fluorescing markings that could be used to ward off enemies. Mazel began to wonder what role fluorescence played on land. Many scientists thought that land-based plants and animals didn’t use fluorescence because fluorescing is inefficient. To fluoresce, a plant must absorb light (which isn’t 100% efficient) and then, of that light, only a fraction of the energy fluoresces. If this process is 10% efficient, a very bright fluorescence is created. It was assumed that reflecting light, which doesn’t require energy,
was widely use to make plants standout. However, the backgrounds of these flowers may tell a different story.
Color
For more of Mazel’s images visit www.glimpsejournal.com
autumn 2009
Mazel is quick to point out that just because something fluoresces, doesn’t mean it has a function. Not all light emission may be utilized—some may just be decoration. But, weeding out the “pollen beacons” from “decorations” is what Mazel plans on investigating next—one backyard garden at a time. w
volume 2.3
The images of a brown grasshopper in the green grass look completely different under blue light. The green grasshopper strongly contrasts with the red stalks. Here, it appears that fluorescence is an accidental byproduct of photosynthesis, which is the process of absorbing energy from sunlight and converting carbon dioxide into energy. The chlorophyll, where photosynthesis occurs, fluoresces red.
REFERENCES Eisner, T. For Love of Insects. 2004. Harvard University Press. Mazel, C. and Fuchs, E. “Contribution of Fluorescence to the spectral signature and perceived color of corals,” American Society of Limnology and Oceanography, 48(1, part 2), 2003, 390-401. Mazel, C., Cronin, T., Caldwell, R. and Marshall, N. “Fluorescent Enhancement of Signaling in a Mantis Shrimp,” Science, January 2, 2004, Vol 303.
(Above) Fluorescence photograph of yellow garden spider (Argiope aurantia). (Right) Fluorescence photograph of grasshopper.
13
Glimpse  
www.glimpsejournal.com
Orderign
colors
by Rolf G. Kuehni
C
A Multifaceted Problem
olor order systems are expressions of the human desire to find order and rules in the natural and perceptual phenomena we encounter. Their development has been complicated by the following
facts: (1) There are a large number of perceptually different colors when comparing samples side by
side; (2) The appearance of a sample can vary distinctly depending on observation conditions. In addition, there is the mystery of the perceptual nature of color itself.
14
1613 Colors, like tastes or smells, are perceptual experiences. Most commonly, they are the result of the interpretation mechanism for certain physical stimuli in our brain/mind. These stimuli are always kinds of lights—either lights we see directly from a source or, more commonly, lights reflected by objects in front of us. Estimates of the number of different color experiences we can have when observing objects vary
widely, ranging from some one million to five million. This diversity of potential experiences indicates the difficulties in defining the issue. How to order these colors in a useful manner has been of interest for centuries. The relationship between color stimuli and experiences is very complex.
Figure 1. Aguilonius’s depiction of color order, with the chromatic primaries: yellow (flavvs), red (rubeus) and blue (caeruleus) enclosed between white (albus) and black (niger). The circular segments represent scales between pairs of the five colors; on top tint/shade scales; on the bottom hue scales. (Aguilon, 1613)
People with normal color vision can have variations in the sensitivity of their sensor types, or cones, in the eye resulting in somewhat different color experiences. Further, tests have shown that individuals vary consider-
ably in their choice of stimuli that, for them, result in the perception of Hering’s unique hues, for example, when yellow is experienced as neither reddish nor greenish.
1704 autumn 2009 Color
These and other phenomena show the impossibility of objectively defining a given color experience. Further, we cannot define color in terms of experience because we are unable to unambiguously and objectively express what it is we experience, nor can we identify with certainty from memory a given previously experienced color. All of these issues make establishing quantitative color scales difficult and problematical. The history of color order systems embodies the growth of understanding of these difficulties. Today, the problem cannot be considered solved in a fundamental manner.
volume 2.3
Phenomena like metamerism, in which stimuli that widely differ in spectral composition of the light produce identical sensory experiences, such as when two paint companies offer the same fashion color comprised of different sets of pigments, resulting in different reflectance curves but identical appearance. Varying visual effects also result from the spectral composition of the light in which objects are viewed. Lights that appear similar can produce widely differing experiences from the same objects. The appearance of artificial daylight produced from a filtered tungsten light and from a three-band fluorescent lamp can be indistinguishable, but certain objects have widely different appearances when viewed in the two lights. In contrast effects, different surround conditions can noticeably affect the experience from a given color stimulus. This is the subject of many colored visual illusions.
Figure 2. Newton’s spectral color circle and mixture diagram, with white O in the center. Orange color Y is diluted with white light to form a saturation scale passing through color Z to the colorless center. (Newton, 1704)
From One-Dimensional Scales to a Riemannian Solid: A Few Highlights
C
olor order began in antiquity with one-dimensional scales. The most famous is Aristotle’s, stemming from his belief that colors are derived from different ratios of white and black. He recognized five major chromatic colors between them, usually translated as yellow, crimson, violet, green and blue. This scale, often slightly modified, remained influential well into the 16th century. In 1613, Flemish mathematician and philosopher Franciscus Aguilonius (1566-1617) reduced the primary colors to three— yellow, red and blue—and added an implicit hue circle passing through orange, purple and green. Tint/shade scales connected the chromatic colors to white and black. There is also an implicit gray scale (Figure 1). As a result, his system is two-dimensional. In the mid-17th century, Isaac Newton (1642-1727) brought a new, scientific basis to color order, which he fully described in his book, Opticks (1704). He recognized seven hues in the spectrum (a number selected to agree with musical scales) and arranged these in a circle, with “white” light in the center (Figure 2). This diagram implicitly describes two color attributes, hue and saturation, the latter varying along radial lines from the center to the periphery, as indicated by the example of orange Y and its desaturated version Z. Newton fully understood a third attribute, brightness, but he did not connect all three in a geometric model.
15
Figure 3. (Right) Mayer’s depiction of the central plane of his double triangular pyramid color solid, with his perceptual primary colors yellow (G), red (R), and blue (B) in the corners, binary mixtures along the periphery, and ternary mixtures in the interior. (Mayer, 1758)
Figure 5. (Opposite left) Sketch of Runge’s color sphere with the primary colors yellow (G), red (R), and blue (B) forming an equilateral triangle on the central plane, with the intermediate hues orange (O), violet (V), and green (Gr), and with white and black on the poles. (Runge, 1810)
1756
Figure 6. (Opposite right) Hand-illuminated image of the hue circle and the gray scale of Grégoire’s color order system with three independent attributes. (Grégoire, ca. 1820)
Glimpse
www.glimpsejournal.com
Figure 4. (Below) Hand-illuminated image of Lambert’s color pyramid. (Lambert, 1772)
16
He also did not leave room for the purple colors that, although not found in the visible spectrum, form an important perceptual hue range. German astronomer Tobias Mayer (1723-1762) took a definite step towards developing a three-dimensional system. In 1756 he described a double triangular pyramid ordering system based on the primary colors yellow, red and blue, plus white and black. Figure 3 shows the central plane of the solid with the chromatic primary colors in the corners of the equilateral triangle and eleven mixed grades between them. The interior of the figure contains mixtures of all three primaries in consistently changing ratios. Mayer’s premature death
prevented him from attempting to color his theoretical system. This task was left to Swiss-German mathematician Johann Heinrich Lambert (1728-1777) together with the painter Benjamin Calau. They quickly encountered problems with Mayer’s system, one being that blackish colors already resulted from mixtures of the pigments used as primaries, yellow, red, and blue. Lambert decided to drop the lower pyramid and reduced the number of grades between the chromatic primaries to seven, resulting in a hand-illuminated pyramid with 164 samples. Figure 4 shows 88 colors of the system. In 1810, German Romantic painter Philipp Otto Runge (1777-1810) pro-
volume 2.3 autumn 2009 Color 17
1810
posed a theoretical system based on the geometric solid of a sphere. He also placed the classical painter’s primaries—yellow, red and blue—in the corners of an equilateral triangle. However, the intermediate mixtures were all taken to have the same saturation as the primaries. He therefore placed the colors on a circle, which along with the tint/shade mixtures, thereby form a sphere with white and black at the poles (Figure 5). The central vertical axis represents a gray scale. The hue plane was considered complementary, that is, consisting of opposing hues that, in a mixture, neutralize each other gradually to form, when balanced, a neutral gray in the center. Again, Runge’s premature death prevented him from attempting
to produce a detailed colored version of the system.
At about the same time, Gaspard Grégoire (1751-1846), a French inventor of a textile art process, introduced the largest color atlas of its time. The French Royal manufacturing companies used the reported 1,351 samples of the atlas for reference purposes. (Unfortunately, copies no longer exist.) The organization of colors within that atlas is unknown, but ca. 1820, Grégoire published a book on color that includes a hand-illuminated description of a small color order system (144 samples) with a cylindrical, color attribute-based organization. Grégoire’s attributes were essentially Newtonian, with a 24-grade hue
(teinte) circle with 12 grades shown, an 8-grade lightness (nuance) scale, and a 9-grade relative saturation (ton) scale (Figure 6). Grégoire attempted to make the steps between grades perceptually equidistant by halving distances between established color samples. Hering’s Perceptual Color Fundamentals
I
n the mid-19th century, James Clerk Maxwell (1831-1879) in England and Hermann von Helmholtz (1821-1894) in Germany began to lay the foundation for the modern psychophysical system, known as the Young-Helmholtz system. It allowed the specification of color stimuli of objects in terms of three
(tri-stimulus) values) that related to the light-reflecting properties of the material, the spectral power distribution of the light in which it was viewed and data that directly or indirectly described the spectral sensitivities of the three cone types of an average human observer. The color stimuli calculated for object colors in this manner fit into an irregular geometric solid, first calculated in the 1920s by German photochemist Robert Luther (1867-1945) (Figure 7). But it was soon found that the relationship between tri-stimulus values and perceptual attributes is a very complex one. Among painters and early thinkers about color, such as Mayer, Lambert, Runge and Grégoire, the idea had solidified that the primary color percepts, a kind of color fundamentals, were yellow, red and blue. But experiments in the late 18th and 19th century indicated that all hues can be matched with many different primary triads, but all more or less limited in the saturation of the achievable color percepts. It later became evident that the best primaries to mix in order to produce all hues at the highest level of saturation are violet, green and red. But there are many possible primary triads. In color monitors, for example, the primary col-
1905
ors in each pixel are red, green and blue (RGB); in color printing, the optimal primaries are cyan, magenta and yellow, with black added for darker colors (CMYK). But what, if any, are the perceptual primary colors, or color fundamentals? In the second half of the 19th century German physiologist Ewald Hering (1834-1918) proposed that there were four chromatic and two achromatic perceptually fundamental colors: yellow, red, blue, green and white and black. The chromatic primaries have unique hues, that is, unique red is simply red, neither yellowish nor bluish (and likewise with the other three unique hues). Fundamental yellow and blue, as well as red and green, oppose each other, forming an “opponent-color system.” Hering argued that one could easily detect yellow and red in orange, but not orange and purple in red (and comparable for the other hues). Figure 8 shows Hering’s construction of the hue circle from different ratios of the four fundamental colors. (Most people today agree with this point of view. Beginning in the 1960s, neurons were identified in the retina and various locations of the brain that represent opponent color processes of sorts. But there
1927
Figure 7. (Opposite below) Two views of Luther’s psychophysical object color solid. (Luther, 1927)
volume 2.3
Figure 8. (Far left) Hering’s conceptual representation of the generation of the perceptual hue circle from the four hue fundamentals. Top: hue circle with amounts, diminishing on both sides of the maximum, of the four fundamental hues; bottom: hue percepts generated at regular intervals from the mixtures on top. (Hering, 1905-1911)
autumn 2009
Figure 9. (Right) Depiction of a constant hue page of the Swedish Natural Colour System. (Scandinavian Colour Institute, 1978, reprinted with permission)
Color
1960 is wide agreement that no valid neurological model of the perception of unique hues yet exists.) Hering stated that any perceived color of an object could be expressed in terms of the content of one or two unique hues plus whiteness and/or blackness, with the sum always adding up to 100%. The hue circle grades represent constant increments of a unique hue with corresponding decrements in the second one, say at 10% intervals. He termed the most highly saturated color of a given hue “full color” and all others “veiled colors.” The full color was placed in one corner of an equilateral triangle, with white and black in the other two corners; the interior filled with veiled colors. Such scaling results in a perceptually evenly spaced system, but the scales are substantially different from those based on
judgments of uniform, equally noticeable differences. Connecting the gray scales of each hue diagram formed a double cone solid that was believed to represent all object colors. The vertical dimension of the solid represents relative lightness, because the full colors form a natural hue circle where different hues have different levels of lightness. It is evident that the only geometrically clearly defined attribute in Hering’s system is hue. Hering named his system the “natural color system.” His concept became important in perceptual psychology around the turn of the 20th century. One hundred years after Hering’s initial description of his perceptual color model, the Scandinavian Colour Institute began to market a modern
19
atlas version under the name Natural Colour System. The present system consists of 1750 samples, arranged on forty constant-hue planes (Figure 9). The claim to “naturalness”—and smart marketing—has resulted in wide popularity of the system. Munsell’s System and the Optical Society of America’s Uniform Color Scales
T
he Munsell Color System represents an approach in which perceptual attributes are clearly in line with geometrical dimensions. American artist and educator Albert Henry Munsell (1858-1918) was interested in a scientifically justified color order system that could be used as an educational tool to develop laws of color harmony. He based it on the perceptual attributes
www.glimpsejournal.com Glimpse 20
1907
isotropic color space would be impossible. The reason is that a solid cannot be filled without gaps with spheres of equal size, each sphere having on its surface colors that are perceptually equally distant in all directions from its central color.
hue, value (lightness), and chroma, modeled in a cylindrical system. A significant shortcoming is its failure to consider the HelmholtzKohlrausch effect describing the fact that perceptual lightness does not just consist of colorimetrically defined lightness, but has an added component that depends on hue. The Munsell hue circle has five primary hues, with a total of forty hues illustrated in the atlas. The vertical value scale has ten grades, and the chroma scale is open-ended. Each scale has perceptually equidistant steps, but they differ in size in the three scales. Figures 10a and b show a schematic view of the system and a view of the abbreviated “Munsell Color Tree.” The first atlas, published in 1907, was updated in 1915 and 1929. In the 1940s, a committee of the Optical Society of America (OSA) generated a large number of new perceptual judgments and proposed revisions to the system, known as “Renotations.” Soon after, these revisions were implemented in the Munsell Book of Color. The Munsell system, consisting in the glossy version of 1550 samples, is today perhaps the most widely used color order system. The Munsell system has three logically well founded, independent attributes in a Euclidean space, but it is not consistently perceptually uniform. In the late 1940s the OSA formed a special subcommittee responsible for the development of a consistently uniform color space (i.e., one perceptually uniform in all directions—an “isotropic” space). Theoretical considerations indicated that to develop a truly
The OSA determined that the geometric solid that can fill a space uniformly with a maximum number of axes along which the differences are uniform is the cubo-octahedron (Figure 11a), with twelve directions of uniformity around the central reference point. The committee obtained difference judgments of seventyseven observers for forty-three color samples, all of equal lightness and presented in triangular arrays of neighboring samples. A psychophysical best-fit model showed substantial discrepancies between the Euclidean geometric model and the average judgments. A key cause is the so-called hue-superimportance effect, already discovered in an earlier mathematical analysis of the Munsell system, In essence, this effect indicates that the human color vision system is particularly sensitive to stimulus differences interpreted as hue differences. Because of this perceptual characteristic, an isotropic color solid cannot be modeled as a Euclidean solid, but rather needs to be formed as a positively curved Riemann solid. The huesuperimportance effect is present in
1940
oday, there are several systems of color order in technical use like in
A form of isotropic color order is represented by psychophysical color difference formulas, technically important for predicting the average perceived color difference between two material samples. The International Commission on Illumination (known as CIE for its French name) has made several recommendations for such formulas since the mid-20th century, the most recent being the formula known as CIEDE2000. These formulas are based on judgments of just noticeable, small differences, but continue to lack a solid experimental basis.
Color
T
Manufacturers of paints or dyes also issue many color order systems. Although these may have a basis in a perceptual system, they are usually biased for commercial reasons and include additional samples of near-grays and special emphases influenced by color trends in fashion.
autumn 2009
Technical and Commercial Color Order and Color Difference Formulas
software such as Adobe Photoshop®: RGB (red, green, blue primaries), HSB (based on approximate perceptual attributes hue, saturation, brightness), CMYK (expressed in terms of the fourcolor printing primaries), and Lab (related to the psychophysical, perceptually approximately uniform, color difference formula CIE 1976 L*a*b*). In addition, several other modern psychophysical color appearance systems, such as CIECAM02, make it possible to adjust for various factors, for example, light source, adaptation and surround conditions.
volume 2.3
all perceptual color difference data, from just noticeable differences to large differences, such as in the OSA data. The subcommittee decided to smooth out the differences so that the results could be shown in a Euclidean space. They also added the HelmholtzKohlrausch effect as well as a second effect involving lightness to the model. They defined 424 color samples, issued in 1977 as the OSA Uniform Color Scales (Figure 11b). The scales constitute a Euclidean system that, of all systems discussed, comes closest to representing an isotropic solid. However, in the system there are no constant hue planes or constant chroma contours. It’s small number of samples and the relative difficulty of orientation in it have limited its commercial success.
Figure 10a. (Far left) Schematic depiction of the construction of the Munsell color order system, with the central value scale, hue circles of various completeness at different values, and radial chroma scales varying by hue. (X-Rite Inc., reprinted by permission) Figure 10b. (Opposite right) Depiction of an abbreviated three-dimensional “Color Tree” formed from constant-hue pages of the Munsell Book of Colors. (X-Rite Inc., reprinted with permission) Figure 11a. (Left) Geometric depiction of 12 samples (small open circles) equidistant from the central sample (black circle), forming the cubo-octahedron solid. Each sample is surrounded by a sphere (heavy solid lines) on the surface of which are all samples with constant difference from the center samples. (Gerstner, 1986, reprinted with permission)
21
Figure 11b. View of a three-dimensional model of the Optical Society of America Uniform Color Scales showing some details of its internal organization (Slide courtesy D. L. MacAdam).
www.glimpsejournal.com
Mathematical equation systems modeling the relationship between physical stimuli and average perceived distances are technologically important but have limitations due to a number of factors.
Glimpse
Developing a universal color order system remains an ideal that cannot be achieved for these various reasons. Chief among them is the complexity and variability of human color perception. w
22
Bibliography 1.
Aguilonius, F. 1613. Opticorum libri sex, Antwerp: Plantin.
2.
Gerstner, K. 1986. The forms of color, Cambridge, MA: MIT Press.
3.
Grégoire, G. ca 1820. Table des couleurs sur trois feuilles, Paris:
1977 Brunot-Labbe.
4.
Hering, E. 1905-1911. Grundzüge der Lehre vom Lichtsinn, Berlin: Springer (English translation: Outlines of a theory of the light
Concluding Remarks
T
he historical record and contemplation of the problem of color organization show that although colors have been ordered in many different ways, a perfect system has not been attained.
sense, L. M. Hurvich and D. Jameson, ed. and transl., 1964, Cambridge, MA: Harvard University Press).
5.
Lambert, J. H. 1772. Beschreibung einer m it dem Calauischen Wachse
Color atlases can be regarded at best as roughly systematic object color stimulus collections, given the variation in color perception by color-normal humans.
ausgemahlten Farbpyramide, Berlin: Haude und Spener. 6.
Luther, R. 1927. Aus dem Gebiet der Farbreiz-Metrik, Zeitschrift für
Two major ordering principles developed in the 19th century remain popular, although both have shortcomings.
A truly isotropic system would offer the largest amount of coherent information but is impossible because of the hue-superimportance effect and the limited number of directions in which uniformity can be expressed.
technische Physik 8:540-558. 7.
colorum commentatio, in Opera inedita Tobiae Mayeri, G. C. Lichtenberg, ed., Göttingen, 1775. 8.
Technical-commercial atlases are often based on commercial, fashion-oriented perceived needs.
Mayer, T. 1758. De affinitate
Newton, I. 1704. Opticks, London: Smith and Walford.
Glossary
volume 2.3 autumn 2009 Color
Achromatic colors: colors without hue, white, black and grays between them. Brightness: an attribute of visual perception that indicates the amount of light emitted by a source or reflected by a surface, ranging from dim to bright. Chromatic colors: colors that have a hue. Chroma: Munsell’s designation of an attribute describing color intensity of object colors. CIE 1976 L*a*b*, CIEDE2000: two “perceptually uniform� psychophysical color difference formulas recommended by the International Commission on Illumination (CIE). CIECAM02: a color appearance formula recommended by CIE that allows prediction of the appearance of a stimulus in a variety of conditions. Color attributes: inherent characteristics of colors, such as hue, lightness, saturation. Euclidean space: a three-dimensional space of classical geometry. Hue: an attribute of color that encompasses its inherent perceptual nature, e.g., the redness of red. Isotropic space: a space uniform in all its properties in all directions. Lightness: perceptual attribute describing the relative amount of light reflected from a surface, relative to that from a white surface, ranging from dark to light. Natural hue circle: a hue circle on a constant plane in which hues are shown in their most common form, regardless of lightness. Neuron: a cell in the nervous system specialized in the transmission and interpretation of electrochemical signals. Psychophysics: the science describing the relationship between material stimuli and human perceptions. Riemannian geometry: a non-Euclidean geometry with positive curvature in which the postulate of parallel lines is replaced by one of every pair of straight lines intersecting, named after the German mathematician Riemann (1826-1866). Saturation: perceptual attribute of color, here generalized to mean intensity of color, ranging from gray to full color. Spectral composition (visual): the composition of a visual stimulus in the spectrum from ca. 400 to ca. 700 nm wavelength of light. Spectral power distribution: the relative amount of light at different wavelengths compared to a reference wavelength. Tint/shade scales: color scales from a full color tinted toward white and shaded toward black. Value: term used by Munsell to describe the attribute of lightness.
23
Glimpse   www.glimpsejournal.com
RetroSpect: 1762 and 1775
24
volume 2.3 autumn 2009 Color 25
Excerpted from Universal Magazine of Knowledge & Pleasure. June 1, 1762. 300-302. In the AAS Historical Periodicals Collection (compilation Š 2009 by the American Antiquarian Society). Printed with permission from the American Antiquarian Society and EBSCO Publishing: http://www.ebscohost.com/thisTopic. php?marketID=1&topicID=1173
26 Glimpse   www.glimpsejournal.com
volume 2.3
autumn 2009 Color
27
www.glimpsejournal.com Glimpse 28
Excerpted from Monthly Review. December 2, 1775. 532-535. In the AAS Historical Periodicals Collection (compilation © 2009 by the American Antiquarian Society). Printed with permission from the American Antiquarian Society and EBSCO Publishing: http://www.ebscohost.com/thisTopic.php?marketID=1&topicID=1173
Color Matters is something entirely engrossing, and yet to me it still remains completely mysterious.
In the time I have taken to investigate color, I have found many different possibilities
qualities. Here, I hope to share some notes and thoughts I have had concerning color; to speak to its inherent unruliness; and also to comment on its open-ended nature as the element to everything.
autumn 2009
for its use. The conditions for it are multiple, a fact that only highlights its infinite
volume 2.3
I
n the past few years I can say that I have had some intense negotiations with color. It
Color 29
30 Glimpse   www.glimpsejournal.com
Color should not behave It should not be subjugated
It should not be submissive
Color should not be quiet,
Color is unruly It is not for the faint of heart. It can be hard It can be strong and bold It can be clear and true It can also lie It can trick and deceive us all. Color does what it wants It misbehaves But most of all, color can change our minds.
(Previous page) Odili Donald Odita, Panic, 2006. Acrylic on canvas, 60 x 80 inches. OD06.016. Image courtesy of Jack Shainman Gallery, New York. (This page, above) Odili Donald Odita, Panoramic, 2009, Acrylic on canvas. (This page, left) Odili Donald Odita, Post Perfect, 2009, Acrylic latex wall paint on wall. From the exhibition, Wallworks at The Yerba Buena Center for the Arts, San Francisco, August 1-October 25, 2009. Photos by Ira Schrank.
Color
nor shy
autumn 2009
It should not play nice.
volume 2.3
It should not obey
31
www.glimpsejournal.com Glimpse   32
Odili Donald Odita, Possibility, 2008. Acrylic on canvas. 84 x 109 inches. OD08.006. Image courtesy of Jack Shainman Gallery, New York.
COLOR QUESTIONS Does a color, seen up close as a shape, become a line in the distance? Where is line when color is a space, or a shape, or a form? Is it the imagined container, the edge, defining this color? Is line the edge created when two or more colors meet?
How does the atmosphere and environment define its key? And what effect does environment have over the color of things in space?
Color
How can we identify the polemics of power through color? Can the way we identify the nuanced balance of power relations between two political bodies be similar to the way we observe the weight and balance of two opposing colors? Can we negotiate the terms of space, and the containment of space, through color?
autumn 2009
Color as a Socio-Political Construct
volume 2.3
What holds color?
33
Color as Culture Can we be defined collectively through the colors we use? And can we be determined through the colors assigned to us? Does geography play into the assignment of color? What of the color of skin? Are we simply “our color?” Are our identities bound by the descriptions of our color? Can our meaning move beyond these basic descriptions? Are we a collection of beings defining and segregating ourselves under the rubric of color? Does color define our communication, our cohabitation, our socialization and our existence? Is it possible to move beyond stereotype and histories created through this aesthetic?
Absolute Color One color is one thing. It cannot adequately be described without placing it next to another color, or without referring to its antithesis, Non-Color. Does anything at all exist without color? What about a blank piece of paper that’s perceived as white? Paint Wite-Out onto that same piece of paper, and ask yourself which is white. Is it the paper, or the Wite-Out? Are they both white? Is one whiter? Is white one thing, or many things? What is the difference? Can we also say that the white paper has now become a soft blue and the Wite-Out a glowing yellow?
www.glimpsejournal.com Glimpse 34
THE COLOR LINE In my work, I’ve explored the psychological condition of color as separate from, and in relation to, the philosophical condition of black and white. I’ve tried to compare black and white to color in ways that speak in terms of cultural conditioning, leading me to consider the complex conditions of African identity within a global context.
looked into aspects of desire/desiring in relation to black and white. And through color, I’ve looked into notions of “the missing” and have spoken about absences identifiable within these spaces. On a psychological level, I’ve examined black and white as a repository for the unfathomable, the unquenched and the unfinished. The tension of incompleteness is found in this type of pictorial space, where the viewer fills in and becomes the void that exists between the polar extremes of black and white.
Black & White The term “black and white” is normally understood as a basic pragmatic idea of clarity within thought, reasoning and presentation. Black and white is also seen as an absolute in terms of value judgment. I‘ve investigated this pretense and have researched, in a deep way, the metaphoric implications “black and white” has within the human consciousness. I’ve
I’ve looked into the ever-persistent problematic condition of black and white as it deals with race—and what manifests as a continually imbalanced state of power between these two absolute value positions.
Color in Line
particular, as well as to culture and the aesthetic. This condition of multiplicity has always been inherent within color; it is now in these contemporary times that we can be freer to discuss these multiplicities without the hinderance of ideological/aesthetic censorship.
The premise of color here is one of description. Color fills in the blank that is left open within a black and white format. Color describes the world in a more complex, if obvious way, and yet the specificity of color can make this newfound complexity that much more alluring and mysterious. Questions Color represents freedom. It is the third position. become even greater in a world of color, as there And it is more than gray. seems to be more to see—and more to choose. I want to engage with color in terms of the Aesthetics, as applied to culture, can define and politics of specificity and difference, and to speak underscore the distinct character of particular toward individuality and a unique distinctiveness histories and societal lines within cultural that runs parallel to the notion of humanity itself. groupings. The reaffirmation of these distinctions within a self can bring outward an understanding of the complexity of relations that weave together varying human experiences. Color in its descriptive state makes a reference to race in Odili Donald Odita, sketches, 2006 and 2008. Courtesy of the artist.
THIRD SPACE
volume 2.3 autumn 2009 Color 35
www.glimpsejournal.com Glimpse   36
Odili Donald Odita, Fusion, 2006, acrylic on canvas, 96 x 120 inches. OD06.010. Image courtesy of Jack Shainman Gallery, New York.
volume 2.3
autumn 2009 Color
37
Chromophobia Chromophobia, the fear of color that exists in the West, is something that runs tangent to the fear of people of color. It is the fear of contamination, of contagion and filth. It is the fear of something different—it is the fear of the unknown.
www.glimpsejournal.com
Challenge of the Unknown
Glimpse
In my paintings, I continue to explore metaphors that address the human condition through pattern, structure and design, and I continue to investigate the possibility of triggering memory. The colors I use in my paintings are personal—they reflect the collection of visions from my local and global travels. Trying to derive the colors intuitively, hand-mixing and coordinating them along the way, is one of the hardest aspects of my work. In my process, I cannot make a color twice—it can only appear to be the same. This knowledge is important to me, as it highlights the specificity of differences that exist in the world of people and things.
38
Fusion What is most interesting to me is the fusion of cultures that takes place where things that seem faraway and disparate function within an almost seamless flow. The fusion I seek is one that can represent a type of living within a world of difference. No matter the discord, I believe that through art, there is a way to weave the different parts into an existent whole, where metaphorically, the notion of a common humanity can be understood as real. Odili Donald Odita Philadelphia, October 2009 w
volume 2.3 autumn 2009 Color 39
(above) Odili Donald Odita, sketch, 2007. Courtesy of the artist. (left) Odili Donald Odita, Eternal, 2007. Acrylic on canvas. 183.5 x 233.5cm. Courtesy of Jack Shainman Gallery, New York,
An Interview with
Glimpse Glimpse
www.glimpsejournal.com www.glimpsejournal.com
Evolutionary Biologist
40 40
Hopi Hoekstra by Carolyn Arcabascio Glimpse Journal (GJ): How did you become interested in the study of color’s evolutionary role in nature?
Hopi Hoekstra (HH): I was initially very interested in studying the genetics of adaptation—the link between genotype (DNA sequences) and phenotype (morphology, physiology, behavior, etc.). In other words, I wanted to know how differences in phenotypes—especially those that matter for how an organism reproduces and survives in the wild—are encoded in the genome. Specifically, I wanted to find precise DNA base pair changes that are connected to variation in fitness in wild populations. This is no small task! To this end, I chose to start with one phenotype: coloration. Color is an ideal phenotype for several reasons. On a strictly practical level, it is easy to measure (we use a spectrophotometer to objectively measure color differences). Another reason is that color can be quite variable, and often differs between closely related species and sometimes even within a single species or population. Moreover, natural variation in color has been well documented and described by early naturalists in many species. Color is a primary way in which organisms interact with their environment—it’s involved in a number of biological processes including mate choice. For instance, flower petals have different colors to attract particular pollinators, which is how plants “mate.” Organisms also use color to issue warnings. Many poisonous taxa, or groups of organisms, are brightly colored to advertise their toxicity. Many non-toxic species mimic the color patterns of toxic species to try to fool predators. Predators or prey can also use color to blend into their environment for protection. This evolutionary tactic is called crypsis. Thus, sometimes even small differences in color can have large effects on an organism’s ability to survive and reproduce in the wild.
We already know a lot about the genes and pathways involved in pigmentation from studies in laboratory mice that have accumulated over the last century. From this work, we also know that the genes/
volume 2.3 autumn 2009 Color 41
pathways are well-conserved across mammals, and even largely across vertebrates. Why all the attention to pigmentation? Because it is a visible phenotype, it has served as a model system for study by geneticists, cell and developmental biologists and evolutionists. In the end, if you are interested in understanding the evolution of diversity, what is a more obvious trait to study than coloration! GJ: You’ve mentioned that the study of color in nature has enjoyed a long history across various scientific disciplines. How has your own research built upon existing knowledge, and what new i n s i g h t s h av e y o u r e a c h e d ?
Courtesy of the Harvard Museum of Natural History. Photograph by Paul Bratescu www.animalexplorer.com.
HH: Our research has built on a wealth of information about animal pigmentation which primarily comes from genetic studies of laboratory mice. Our work has focused on taking advantage of this information to understand the genes and developmental
42 Glimpse   www.glimpsejournal.com
(Left): Courtesy of the Harvard Museum of Natural History. Photograph by Carla Richmond, www.go-carla-go.com/www.hmnh. harvard.edu. (Below) Courtesy of the Harvard Museum of Natural History. Photograph by Adam Blanchette, www.hmnh.harvard.edu.
GJ: What are some implications of this convergence involving the melanocortin-1 receptor? Now that you’ve linked the expression and evolution of color pigmentation to the same gene across species, is this telling of the broader genetic relationship between species? volume 2.3 autumn 2009
HH: Mc1r seems to be responsible for color variation (especially melanism, or completely black pigmentation) in a number of different vertebrate species ranging from mammals to birds to lizards. To an evolutionary biologist, this pattern suggests several interesting things: (1) that the pigmentation pathway is largely conserved across vertebrates (that is, Mc1r mutations cause the same types of changes in color in a mouse fur, bird feathers and a lizard scales), and (2) that evolution, in some sense, may be predictable since the same gene is repeatedly targeted in very43 diverse species. Color
process responsible for natural variation, either by studying wild populations or by studying color patterns that just don’t exist in lab mouse colonies (like stripes and spots). We are interested in questions about how natural selection acts on pigmentation genes to produce that amazing diversity of color and color patterns that we see out there in natural world. We have made some very interesting, and sometimes surprising, discoveries. Here is one: pale lizards that live on the sand dunes of White Sands National Monument in New Mexico have mutations in the same gene as lightcolored mice inhabiting Florida’s Gulf Coast beaches. This commonality is what we refer to as a convergence, in which distantly related species independently evolve similar traits because they live in the same habitat, and in this case, they do so by genetic changes in the same gene! This gene, the melanocortin-1 receptor, is also known to affect color in a wide variety of vertebrates, from melanic jaguars to snow geese to red-headed humans. This one gene is thus responsible for a large amount of color diversity in nature.
Why, you may wonder, is Mc1r so often involved? There are several potential explanations. First, it is a large gene so it represents a big “target” for mutations. And there is some researcher bias because this gene, and its conserved structure, makes it easy to examine in a number of diverse species; it is tested more than other pigmentation genes. But the most biologically interesting and relevant explanation is that Mc1r is quite specific to pigmentation. Mutations in this gene cause color change but they don’t mess up anything else, so the net selection advantage is high. GJ: Could you give some examples in which a species’ color mutation has been advantageous from a survival standpoint?
HH: There are several ways that color can improve fitness, the differential survival or reproduction of individuals. Perhaps the most straightforward example, and one that comes from our own research, is camouflaging color in mice. There is a very widespread pattern in rodents that their dorsal coats, or the fur on their backs, often closely match the substrate
for color differences in snow geese and arctic skuas— differences that matter for mate choice. In snow geese, females tend to choose males that have color patterns similar to hers.
These same genes that influence that coloration of mice also contribute to human skin coloration, which is also an adaptation. It is thought that heavily pigmented skin helps protect against the sun’s cancerous effects, hence dark skin is beneficial near the equator. However, the sun’s UV rays are also important in Vitamin D synthesis, so where the sun’s rays shine less bright near the poles, the skin is less pigmented to allow for this process to happen. In the North, then, too much pigmentation may lead to Vitamin D deficiency.
These are cases in which we know something about the genes. There are many examples of color and color patterns that are thought to improve survival—like zebra stripes and leopard spots—for which we don’t know (yet) what the genes are.
www.glimpsejournal.com
which they in habit. So, we have been examining areas with extreme soil colors—from lava flows to white sand dunes—to see how the mice have adapted and to understand the genetic basis of this adaptation
Glimpse
In addition, color can be important in attracting mates. This is often true in birds. Work by Nick Mundy’s group at Cambridge University has shown that MC1r is responsible
44
GJ: What is the next step for your research?
HH: In terms of pigmentation research, we are starting a series of new projects to answer the question of how regular color patterns are made (How did the leopard get his spots?). Because conducting experiments with leopards, tigers or zebras is impractical, we are working with smaller and more abundant species like chipmunks and thirteen-lined ground squirrels, which have regular striping patterns that presumably aid in camouflage. Learning about the genes and developmental processes that produce more complex color patterns will represent a big step forward in our understanding of the evolution of biological diversity. GJ: Is the way that color is distributed in pattern among vertebrates also determined by the melanocortin-1 receptor?
HH: One can think of Mc1r as being the switch that tells a pigment-producing cell (melanocyte) whether to produce dark (eumelanin) or light (phaeomelanin). In striped animals, it is possible then that the switch is in a different position—”on” in the areas of dark stripes, and “off” in areas of light stripes. However, what we are really interested in is what genetic or developmental change flips the switch in one direction or other. Most likely, something upstream in the pigmentation pathway is differentially expressed during development, and this lays down a type of “map” on the developing organism, laying down the plan where the stripes or spots will eventually be. Once we have a better idea of how these complex patterns are made in vertebrates, it will be fun to see if they do it in the same or different ways in invertebrates. There are several research groups trying to understand the genetics and development of butterfly wing patterns, for example. It is a really exciting time to be studying the evolution of color and patterns in nature!
volume 2.3 autumn 2009 Color 45
G J : W h a t i s y o u r f av o r i t e c o l o r ?
HH: My favorite color is blue. Funny enough, this color is very rare among mammals—limited to just a few parts of just a few species such as the rump of a male mandrill. Moreover, blue isn’t generated by simple pigments like red or yellow color, but is instead the result of light reflecting off of small microstructures; it is a structural color. Blue is unique in this way.
(Opposite) Courtesy of the Harvard Museum of Natural History. Photograph of a Dart Frog by Brian Gratwicke. (Top) Courtesy of the Harvard Museum of Natural History. Photograph of a Panther Chameleon tail by Paul Bratescu, www.AnimalExplorer.com. (Above) Courtesy of the Harvard Museum of Natural History.
But that’s not why I like blue. Blue is the color of the ocean, and when I am not at work there is no place I would rather be than on the water. w Visit www.glimpsejournal.com for details of the exhibit, which runs until March 2010, and for a video of Dr. Hoekstra discussing the exhibit.
www.glimpsejournal.com Glimpse   46
Figure 2A, B, C. (A) Crayon colors filtered through the lens of a young eye. (B) Crayon colors filtered through the lens of an aging eye. (C) Perceived color to the aging lens. Image courtesy of Chris Metcalf, www.flickr.com/photos/ laffy4k/404313786/
Seeing Red on Mars volume 2.3
Adaptation and the influence of the environment on color appearance by Michael A. Webster
autumn 2009
M
The impression of colors as generally stable properties of objects hides the fact that the visual coding of color is a highly dynamic process, for the visual system is constantly recalibrating whenever the stimulus before us changes. In fact, these recalibrations help maintain stability, or color constancy, by allowing us to have the same experience of a surface even when the lighting, and thus the light reflected to the eye, changes.2 For example, a sheet of paper looks white whether we see it in dim or bright light, even though the intensities reaching the eye may differ greatly. The same adjustments can also maintain constancy as the observer changes. As we age, the lens of the eye becomes progressively more yellow, screening
Figure 1. A modern color system that arranges colors like a planet. Different hues circle the equator and become darker or lighter as they reach the poles. Courtesy of the author.
more of the short wavelengths from reaching the receptors. Yet as we ourselves gray, the world does not appear to yellow, because the visual code for gray is constantly renormalized to match the environment (Figure 2).3 Skiers often notice a hint of these effects as they adjust to their yellow goggles and must then readjust to a bluish-looking world when their goggles are removed. Natural changes in the lens are instead very gradual, but cataract patients afford a natural experiment for exploring these processes when their lens is abruptly replaced in surgery. A blue aftereffect can persist for several weeks before their color perception reapproaches how they experienced white before surgery.4 One of the principal ways that the visual system achieves these adjustments is through sensory adaptation.5 Like cameras, neurons in your visual system must be properly “exposed” to register the picture. If sensitivity is too low or too high, the neurons will either not respond or saturate. Adaptation rescales responses so that they are centered on the average stimulus level. Color is initially registered by the responses in three types of photoreceptors, each with a different sensitivity to wavelength. Light adaptation within each receptor class will adjust their sensitivities so that their responses are equated for the average stimulation— even if that stimulation is strongly biased in its wavelength composition. This balanced response is probably the earliest neural correlate of “gray.”6
Color
odern color order systems usually arrange the palette of colors we experience like a planet with different hues circling the equator and with the poles corresponding to dark and light (Figure 1). At the center of this world is gray, and all other colors owe their identity to how they differ from this norm— in strength (saturation) and kind (hue and hue/lightness ratio). But to what standard does gray owe its identity? The answer cannot lie entirely within us. Our brains have too many variables to come standardized and factory-set to work correctly, and in any case, a brain designed for one situation would not work well in another. The visual system solves this problem by adjusting how it works to try to match the situation it is currently in. Thus how we see colors depends fundamentally on the color characteristics of our environment.1 What looks gray may be the perceptual correlate of the average spectrum of your world, and whether you and I see the same surface as gray thus depends on whether we live in the same or different visual worlds.
47
www.glimpsejournal.com Glimpse   48
Figure 3. Courtesy of the author.
We can experience these adjustments by fixating for several seconds between the four colored squares in Figure 3, and then shifting to the center of the four blank squares. The colored afterimages arise because at each location, the eye is adjusting the white balance according to the current average stimulus, so that the stimulus that previously corresponded to white (in the blank squares) is shifted toward the opposite or complementary color.
volume 2.3 autumn 2009 Color
The same processes that maintain color appearance can also lead to changes in appearance when the color of the world itself changes.7, 8 And the world does change—in different places and at different times. Figure 4 shows roughly the same scene in India during the monsoon and dry seasons. As the landscape cycles from lush to arid, an individual living there would be exposed to very different distributions of color, and their color perception may cycle along with the landLike cameras, neurons must also adjust their response to contrast scape as they adapt to these changes.9To the extent so that they can properly register the variations in stimulus levels that we understand these adaptation effects, we can around the average. That means turning up or down the incremen- simulate how the world should appear to individuals tal response as the stimulus gamut shrinks or expands. It is because living in different environments.10 This requires knowledge of the physical color statistics of different environments, and plausible models of how color information is represented and adapted by neurons in the visual system. But we now have sufficient understanding of human color vision to make reasonable guesses. The bottom two images in Figure 4 were generated by an empirically-based model of color coding to predict how the images would look to an observer completely adapted to the colors characterizing either season. Specifically, the images are rendered after adapting the receptors and the contrastcoding cells so that for each Figure 4. Juricevic, I.,Webster , M. A. (2009). Variations in normal color vision. V. Simulations of adaptation to simulated neuron the avernatural color environments. Visual Neuroscience 26, 133-145. Figure 1. Reprinted with permission. age response is the same within either season. of these adjustments that the perceptual space of colors is more rounded like a planet, for the space of receptor signals is more Note that one effect of the adaptation is to tone down misshapen like an asteroid. For example, because the receptors the colors that are more common (e.g. greens during have overlapping sensitivity to wavelengths, the differences the monsoon and yellows during the dry period) while between them (which signal color) are many times smaller than the highlighting colors that are more rare (yellows in mondifferences within them (which signal brightness). Yet the world soon and greens in dry). This fits well with many peodoes not look like it varies more in brightness than in color, because ple’s anecdotal experiences. Residents of Nevada can the neural response to luminance or chromatic contrast is again wax poetic about the faintest blush of green in spring adapted so that each matches the range of available contrasts. For hillsides, while tourists find the same hills dismally color, these adjustments occur in cells later in the visual pathway, brown. The tendency of adaptation to increase the in populations of neurons in the visual cortex which are tuned to salience of novel colors suggests that another advandifferent hue angles, and in different combinations of luminance tage of adaptation is to build a prediction about the and color contrast. world, in part to emphasize what is new or unpre-
49
www.glimpsejournal.com Glimpse   50
Figure 5A. (Above) How the sea looks to a human from land. Figure 5B. (Below) How the sea may look to fish. Image of courtesy of TANAKA Juuyoh, www.flickr.com/photos/tanaka_juuyoh
dicted.11 This means that the colors that draw your attention are the very ones you have not adapted to. In this sense, much of what you notice about the world may be a visual aftereffect.12
Note also that while vision underwater may seem like an artificial example, the changes predicted may be no larger than the changes we face when we initially adapt to the world. Indeed Figure 6A. (Left) How Mars looks from Earth. 6B. (Right) How Earth looks to “Martians.” ,the bulk of the work done by adapta- Figure Image courtesy of NASA, www.nasa.gov. tion may initially occur when the visual system calibrates itself during development, and the fine tuning that occurs across the should tend to be normalized in the same way if they live in a comrest of the lifespan may often be small by compari- mon world, and should thus have convergent perceptual experison (However, large changes can also result from ences. Conversely, even physiologically identical observers will visual disease. How adaptation might partially com- have divergent perceptual norms if they occupy different worlds. pensate perceptual experience for losses in sensitiv- Note that in both cases, it is the stimulus that determines the ity with disease remains largely unknown.) perception—for it is the mind that is fit to the world14—and this means that some private subjective experiences may be directly This example is also relevant because we increas- traceable to physical and objective properties of the world. To ask ingly explore, live in, and are adapted by new color whether two individuals experience the same stimulus as gray, we environments (many of our own creation) that our may not need to measure the observers, for we can measure the species has never experienced before. The arbitrary color statistics of their worlds.
Color
As these illustrations suggest, the links between the perceptual and physical world through adaptation provide a possible window into the problem of other minds.12 If adaptation behaves similarly in different brains, then we can meaningfully predict some aspects of how color is experienced by others. Of course there are many aspects we cannot predict—including the qualia associated with most stimuli (e.g. whether what feels “red” to me feels “green” to you). But how color perception is normalized—what feels neutral to you—may strongly depend on what you are adapted to. Even observers with intrinsic physiological differences (e.g. old and young eyes)
autumn 2009
either as we might see it with our terrestrial eyes, or as an observer might see it who actually lives in—and is adapted to—the marine environment. Again, this was simulated by adjusting the sensitivity of the modeled color neurons so that their average response to the underwater distribution was the same as the average responses to an ensemble of outdoor land scenes. Note that this makes it easy to see objects in the scene that are barely discernable under the wrong state of adaptation. Thus it is clear in such cases that adaptation helps you to see better, a point which most vision scientists assume but which has often been difficult to demonstrate by experimentally adapting observers, perhaps because we cannot keep the observers long enough to see the full effects of adaptation emerge.13
volume 2.3
To visualize the full power of adaptation, it is helpful to consider more extreme changes in the environment. Figure 5 shows an underwater image
colors that make up the modern world are very different from the colors of the past. This raises the question of how our color percepts differ from our ancestors. What might ancient art have looked like through eyes adapted to the ancient world? And how will things look in the future? When our descendants colonize the Red Planet, the planet may not look red to them, and instead may appear to them much like a landscape we could encounter on Earth (Figure 6).
51
identity
Glimpse
www.glimpsejournal.com
strength
52
Figure 7. (Above) Facial characteristics, like color characteristics, draw more attentions the further they are from the norm. Image courtesy of the author. Figure 8. (Below) Image courtesy of the author.
The average face in a villiage of...
narrow faces neutral faces wide faces
How faces appear to someone living in a villiage of...
narrow faces
neutral faces
wide faces
A final point to emphasize is that these effects are not limited to color; indeed, the visual system seems to adapt to most of the characteristics we can perceive in the visual environment, and color may provide the paradigmatic example for understanding perceptual adaptation more generally. For example, studies of human face recognition suggest that individual faces are encoded by their “identity trajectories” relative to an average or prototype face (analogous to “gray” in color).15 Consistent with this, face caricatures work by exaggerating how the individual’s features differ from the average,16 and are formally similar to increasing saturation to emphasize the hue that defines an individual color (Figure 7). Like color, the perception of faces is also highly susceptible to adaptation. 17 As we view and thus adapt to a particular face, that individual tends to appear more neutral or “normal,” and this induces corresponding aftereffects in the appearance of other faces. If we lived in a world where the average face tended to have more contracted or expanded features, then those features would look normal and possibly even identical to the average face seen by an observer living in a different social world (Figure 8). This hypothetical example becomes more pertinent when considering that again like color, the characteristics of faces naturally vary in different places and contexts—so that we are all exposed to different distributions of identities, ethnicities, ages and genders, all further distorted by individual and cultural differences in fashion. Research on face adaptation has shown that we do adapt to these differences.15, 18 One can imagine how the appearance of faces changed among small confined groups—as a party of immigrants set sail for new lands—or how it might change when a new generation of immigrants embarks for the stars. How will we and the earth we inhabit look to them when they return? w
References Webster, M.A. (2009). Calibrating color vision. Curr Biol 19, R150-152.
2.
Smithson, H.E. (2005). Sensory, computational and cognitive components of human colour constancy. Philos Trans R Soc Lond B Biol Sci 360, 1329-1346.
4.
Delahunt, P.B., Webster, M.A., Ma, L., and Werner, J.S. (2004). Long-term renormalization of chromatic mechanisms following cataract surgery. Vis Neurosci 21, 301-307.
5.
Webster, M.A. (1996). Human colour perception and its adaptation. Network: Computation in Neural Systems 7, 587-634.
6.
Webster, M.A., and Leonard, D. (2008). Adaptation and perceptual norms in color vision. J Opt Soc Am A Opt Image Sci Vis 25, 2817-2825.
7.
Webster, M.A., and Mollon, J.D. (1997). Adaptation and the color statistics of natural images. Vision Res 37, 3283-3298.
8.
Webster, M.A., Mizokami, Y., and Webster, S.M. (2007). Seasonal variations in the color statistics of natural images. Network 18, 213-233.
9.
Juricevic, I., and Webster, M.A. (2009). Variations in normal color vision. V. Simulations of adaptation to natural color environments. Vis Neurosci 26, 133-145.
10. McDermott, K., Juricevic, I., Bebis, G., and Webster, M.A. (2008). Adapting images to observers. In Human Vision and Electronic Imaging, SPIE, Volume 68060, B.E. Rogowitz and T.N. Pappas, eds., pp. V-1-10.
13. Clifford, C.W., Webster, M.A., Stanley, G.B., Stocker, A.A., Kohn, A., Sharpee, T.O., and Schwartz, O. (2007). Visual adaptation: neural, psychological and computational aspects. Vision Res 47, 3125-3131.
Color
Werner, J.S., and Schefrin, B.E. (1993). Loci of achromatic points throughout the life span. J Opt Soc Am A 10, 1509-1516.
12. Webster, M.A., Werner, J.S., and Field, D.J. (2005). Adaptation and the phenomenology of perception. In Fitting the Mind to the World: Adaptation and Aftereffects in High-Level Vision, Advances in Visual Cognition Series, Volume 2, C. Clifford and G. Rhodes, eds. (Oxford: Oxford University Press), pp. 241277.
autumn 2009
3.
11. Barlow, H.B. (1997). The knowledge used in vision and where it comes from. Philos Trans R Soc Lond B Biol Sci 352, 1141-1147.
volume 2.3
1.
53 14. Clifford, C., and Rhodes, G. (2005). Fitting the Mind to the World: Adaptation and Aftereffects in High-Level Vision, Advances in Visual Cognition Series, Volume 2 (Oxford: Oxford University Press). 15. Leopold, D.A., O’Toole, A.J., Vetter, T., and Blanz, V. (2001). Prototype-referenced shape encoding revealed by high-level aftereffects. Nat Neurosci 4, 89-94. 16. Lee, K., Byatt, G., and Rhodes, G. (2000). Caricature effects, distinctiveness, and identification: testing the face-space framework. Psychol Sci 11, 379-385. 17. Webster, M.A., and MacLin, O.H. (1999). Figural aftereffects in the perception of faces. Psychon Bull Rev 6, 647-653. 18. Webster, M.A., Kaping, D., Mizokami, Y., and Duhamel, P. (2004). Adaptation to natural facial categories. Nature 428, 557-561.
WAVEs of Co An ecological valence theory of human color preference An interview with Karen B. Schloss and Stephen E. Palmer
Glimpse Journal (GJ): I understand that y o u t w o h av e b e e n s t u d y i n g p e o p l e ’ s color preferences for the last few years. What’s the focus of your research? Karen Schloss (KS): We’re interested in just about everything about people’s color preferences, but our primary interest these days is finding out why people like the colors they do—and, for that matter, why they have color preferences at all! Stephen Palmer (SP): It turns out that scientists don’t know much about such things, even though they’ve been studying color preferences for almost a century. We have an interesting new theory that we think sheds new light on these questions.
GJ: That seems to imply that there are at least some older theories. What are they? KS: One is that color preferences come from feelings that colors evoke when people look at them. Ou and colleagues call them color emotions.1,2 They found that people generally like colors that feel active, cool and light more than they like colors that feel passive, warm and heavy. SP: Another theory is that color preferences have been hardwired into the physiology of the human visual system by evolutionary factors. We don’t have time to go into the details here, but Hurlbert and Ling found that females slightly preferred reddish colors to blue-greenish ones, whereas males slightly preferred the opposite.3 They argued that that this gender difference evolved in ancient “hunter-gatherer” societies. They claim that the visual systems of the “gatherer” females became specialized for finding red berries amongst green foliage, leading to higher preference for reds than greens. Both of these ideas—color emotions and hardwired preferences—are interesting and have merit, but neither of them held up very well when we tested them against our own data.
Image courtesy of flickr member: Ajar
It may be easiest to understand this idea if you think about color preferences as being analogous to taste preferences. People don’t have sensory access to the nutritional content of food, but we use taste experiences as a surrogate for that information. They steer us toward eating healthy stuff and away from eating harmful stuff, and that’s a good thing.
GJ: That sounds pretty plausible, but w h a t s o r t o f e v i d e n c e h av e y o u f o u n d to support this idea? SP: We’re currently collecting data to test the theory by studying the relation between people’s average color preferences and their average object preferences. If the theory is right, then people’s preference for a given color should be highly correlated with their degree of attraction to the objects they associate with that color. The experiment is sort of complicated because it requires collecting data from three different tasks, but its results, so far, are providing strong support for our theory.
Color
SP: We suggest a new, evolution-based answer that we call the Ecological Valence Theory, or the EVT for short. It actually unites these two previous theoretical ideas and extends them in interesting ways. In a nut shell, we propose that the average preference for a given color is determined by people’s average preference for all the objects associated with that color—how much people in general like or dislike objects of that color. The EVT implies that people should like colors that are associated with things they like, such as blues with clear sky and clean water, and dislike colors associated with things they dislike, such as browns with feces and rotting food. The EVT follows an evolutionary argument to its logical conclusion, because it claims that organisms will be better equipped to survive and reproduce in their ecological niche if they are attracted to colors associated with positive, healthy things and avoid colors associated with negative, unhealthy things. The basic premise of the EVT is that color preferences work by steering people’s behavior in evolutionarily advantageous ways.
KS: For instance, most people are attracted to foods that taste sweet because sweetness correlates with high caloric content in nature, and consuming enough calories is necessary for survival. On the other hand, we tend to dislike foods that taste bitter because bitterness correlates with toxicity, and that’s harmful to survival. We propose that a similar ecological steering function will work for colors, provided that how “good” versus “bad” colors look to us correlates with how evolutionarily advantageous versus disadvantageous objects of that color are to us, at least on average. According to the EVT, color preferences could be determined both by hardwired biological mechanisms (such as Hurlbert and Ling’s cone-contrast based theory) and by emotional associations (such as Ou’s coloremotion theory) that are learned throughout an organism’s lifetime. That’s why Steve said that the EVT unites these two different approaches within a single theoretical framework.
autumn 2009
GJ: So what is your new theory, and how is it different from these other theories?
volume 2.3
lor
55
www.glimpsejournal.com Glimpse  
L
Lightness
O
Y
H
56
Hue
R
G
Saturation
P
B
C
D
Figure 1. A perceptual color space showing red (R), yellow (Y), green (G), blue (B), orange (O), chartreuse (H), cyan (C) and purple (P). The hues are sampled at: a saturated (S) cut, a muted (M) cut, a light (L) cut, and a dark (D) cut. Courtesy of the authors.
M S
GJ: And what are these three tasks?
KS: We selected our colors within this perceptual color space by first choosing the perceptually “unique” hues of red, yellow, green and blue. Then we chose the four hues that contain approximately equal amounts of the unique hues: orange, chartreuse (yellow-green), cyan (blue-green) and purple. Then we sampled these eight hues at four different “cuts” through color space: a saturated cut, consisting of the most “vivid” colors of each hue, a muted or “desaturated” cut midway between saturated cut and the central gray axis, a light or “pastel” cut at the same saturation level but lighter than the muted cut, and a dark cut that was also equal in saturation with the muted cut but darker.
GJ: What do you think could account for these different preferences of dark colors? SP: We think they are different because of how people feel about the kinds of objects that characteristically have those colors. The least preferred colors, which are browns and olives, are the colors of feces, vomit, rotten food and pus. The most preferred color—saturated blue—is the color associated with sunny skies and clear water. These examples lend intuitive support to the EVT, but they aren’t systematic enough to test it. KS: To do that rigorously, we need to consider all the other objects people know about that are brown, olive and blue—and all the objects that are associated with the other colors we studied. That means we had to collect a truly representative set of color-associated objects and then measure people’s average preferences for all of them. These are the second and third tasks of our study, which together we call the Weighted Affective Valence Estimate procedure, or WAVE for short.
Image courtesy of flickr member: Mish Bradley
Color
SP: To understand why we chose the colors we did, you first have to understand the notion of a perceptual color space. This is how color scientists specify colors. A perceptual color space is a model of color appearance in which each color is identified with a point in a three-dimensional space. The locations of colors within the space are determined by color similarities, so that similar colors are closer together in the space and dissimilar colors are farther apart. The three dimensions of human color space are hue (the basic experiential quality of the color), saturation (the vividness of the color), and lightness (the extent to which a color is light vs. dark). The hue dimension is often further broken down into two components: how red versus green the color is and how yellow versus blue it is.
people generally like saturated colors somewhat more than light and muted colors, their relative preferences for hues within these cuts are essentially the same: they like blues most and chartreuses least. But, dark colors seem to be quite different. People in general really dislike dark orange (which is brown), and dark yellow (which is olive), and they like dark red much more than the other reds we tested.
autumn 2009
GJ: You said you’re using thirty-two chromatic colors. What are they, and how did you choose them?
KS: Well, we haven’t finished collecting all of the data yet, but here’s what we’re finding so far. In color preference ratings, there’s a broad peak around blue and a trough around chartreuse. This is similar to what other researchers have reported previously. We’re also finding that even though
volume 2.3
KS: First we get one group of people to rate how much they like each of the thirty-two chromatic colors we’re studying. The second task is to have a different group of people name as many objects as they can that each of the same set of thirty-two colors reminds them of. The third task is to have another different group of people rate their preference for the objects that were named in the second task. They key question is: What is the relation between the average color preference ratings in the first task and the average object preference ratings for the objects in the third task?
GJ: What did you find in the results of your experiment?
57
Glimpse
www.glimpsejournal.com
G J : A n d w h a t h av e y o u f o u n d w h e n you’ve broadened the color objects for consideration?
58
KS: Our preliminary data show that the correlation between the WAVE data measuring preferences for objects of a given color and color preferences for the corresponding color is remarkably high: greater than 0.85! This means that people’s average color preferences are very closely related to their average preferences for objects of that color. And this is what the EVT predicts. People seldom seem to have any idea that their feelings about colored objects might be driving their color preferences when they make their ratings. They just rate the colored squares we show them in terms of how much they like the colors they see. GJ: Interesting. But correlation isn’t causation, is it? How can you know that the high correlation isn’t because people’s color preferences determine their object preferences? Maybe people just like clear, sunny skies because they like blue? SP: That’s a really important question! We think the answer comes from cases in which objects of almost the same color receive completely opposite valence ratings. Your suggestion implies that people dislike feces because it’s brown. If that’s so, why don’t people dislike chocolate, which is also brown? GJ: Point taken. But clearly there are cases in which we like certain objects specifically because we like their color. I like the blouse I’m wearing right now, for example, mainly because of its color. Could that be what’s giving you the results you’re finding? KS: There are a couple of answers to that. One is that when we do the object-naming task in the WAVE procedure, we specifically eliminate descriptions of objects that could be any color, such as blouses, T-shirts, crayons and things like that. This means that those kinds of
Image courtesy of flickr member: Peter Kaminski
examples aren’t actually in our data set at all. The other answer is that we don’t claim that color preferences can’t influence object preferences. Clearly they must for objects like crayons that differ only in color. But color can dominate other factors even for blouses. If you’re shopping for a blouse among many that are roughly similar except for their color, your color preferences probably determine which blouse you decide to buy, because the other differences aren’t as important to you as the color differences. The EVT suggests that if you buy the blouse because of its color, and if you enjoy wearing it, this experience will reinforce your preference for its color. That’s interesting because it means that color preferences can be self-fulfilling prophecies, especially when color is the most important issue at stake. GJ: As I understand it, the evidence supporting your ecological valence t h e o r y i s b a s e d o n av e r a g e c o l o r preferences. But different people can like wildly different colors, can’t they? How can your theory account for this variability? SP: You’re absolutely right. There certainly are huge individual differences in color preferences. For each person who absolutely loves a given color, there is another who absolutely hates it. The EVT can potentially explain such differences as long as different people have different emotional reactions to colored objects. There are two ways this can happen. First, different people might have different object associations for the same color. Brown might
Images courtesy of (left to right) Flickr members: Robert Donovan, Wagner T. Cassimiro & Everjean remind me of feces much more than chocolate, whereas brown might remind you of chocolate much more than feces. The other is that different people might have different emotional reactions to the same objects. Many people love chocolate, but there are others who actually dislike it. This is how the EVT would explain the variation we see in color preferences across people. It also suggests that there will be lots of factors that influence color preference at different levels. Some will cause people’s color preferences to be similar to each other’s and other factors will cause them to be different from each other’s. GJ: By “similarities” you mean that virtually everybody likes clear sky and virtually nobody likes feces, right? KS: Right. Emotional reactions to some objects are nearly universal. Effects of this type could even be innate, meaning that they are selected genetically across generations through natural selection. They could also reflect individual learning through common emotional reactions to environmental objects. G J : S o w h a t a re s o m e o f t h e fa c t o r s that might cause differences in color preferences? SP: One factor is cultural differences. We’re currently collecting data with collaborators in Tokyo, Japan. The data so far suggest that Japanese participants like the light (pastel) colors in our sample more than their American counterparts do, presumably because of cul-
tural factors. The EVT implies that this occurs because the Japanese participants view whatever objects they associate with these light colors as having a more positive emotional valence, on average, than American participants do. Such cross-cultural effects could arise from either or both of two sources. First, people in different cultures may have different objects associated with the same colors. For instance, few Americans would mention “Shinto shrines” as an associate for saturated red-orange, whereas many Japanese people would. Second, people in different cultures may have different valences for the same objects. For example, Americans may not like rice nearly as much as the Japanese do. Together, such differences may account for cultural differences in color preferences. The crucial test for the EVT is whether the WAVE data produced by the Japanese correlate more strongly with Japanese color preferences than with American color preferences, and whether the WAVE functions produced by the Americans correlate more strongly with American color preferences than with Japanese color preferences. KS: Another level of potentially relevant factors comes from sub-cultural social influences related to special-interest groups and institutions, like sports teams, universities, religions and gangs. Strong social ties like these are relevant for the EVT because social institutions also affect people’s success in thriving and producing offspring in modern societies. Such affiliations can have powerful influences on a person’s employment and mating oppor-
www.glimpsejournal.com Glimpse 60
Image courtesy of Flickr member: Anne Oeldorf-Hirsch
tunities, both of which are incredibly important to their lives. To take perhaps the most extreme case, members of an urban street gang are more likely to survive and reproduce if they like and wear their own gang’s colors and dislike and don’t wear the rival gang’s colors. SP: More generally, the EVT predicts that if a group of people is positively invested in a specific social group that’s strongly identified with specific colors, then those people should have elevated preferences for those colors. If other people have negative associations with the same organization, they should have depressed preferences for its colors. We’re actually about to test this prediction, not with street gangs, but with preference for university colors in students that have high versus low amounts of school spirit in athletic competitions. Image courtesy of flickr member: Shaire Productions
G J : Y o u h av e n ’ t y e t a c c o u n t e d f o r t h e differences at the level of single individuals. Why do I like the color of this blouse? It’s my f av o r i t e c o l o r a n d i t i s n ’ t a s s o c i a t e d w i t h a n y social institution I know of. KS: You’re right, of course. There are tons of idiosyncratic factors that are probably relevant to understanding color preferences for individuals. Everyone has unique experiences that may influence those preferences. For example, suppose you used to go visit your grandmother every summer as a child, and she had a bright pink armchair in her living room where she always sat after dinner. If you cherished your visits with Grandma and loved sitting in her lap in the pink chair, then you may well have an increased preference for bright pink relative to the rest of the population. If you loathed your visits to your dreaded grandmother, however, especially when she made you sit in her lap in the chair, you may well have a decreased preference for bright pink relative to the rest of the population. Like for cultural differences, the EVT predicts that an individual’s WAVE data should predict his or her color preferences better than other people’s WAVE data would. Testing this prediction won’t be easy, but we’re about to begin trying in our laboratory.
G J : G i v e n w h a t y o u ’ v e d i s c o v e re d s o fa r , w h a t would you say is the take-home message from your research?
volume 2.3
SP: That the EVT offers a compelling explanation not only for why people, in general, prefer some colors more than others, but also for why we have color preferences at all. People like the colors that bias them toward objects and events with which they are likely to have positive experiences and away from objects and events with which they are likely to have negative experiences. These associations between the colors of objects and the consequences of interacting with them could arise from genetic adap-
autumn 2009
tations resulting from evolutionary selection, from associative learning that takes place during an organism’s lifetime, or from both together. When you approach color preferences with this perspective, they appear to be biologically adaptive mechanisms, helping individuals seek out objects and events that are beneficial for them both physically, such as finding fresh food, and emotionally, such as bonding with others in valued social groups.
Color
KS: They’re also at work, of course, in influencing people’s choices among artifacts that vary only, or mainly in color, such as clothes, cars and iPods—as market researchers know all too well. Here, the value of color preferences also lies in helping us make beneficial choices, but it’s because the alternative objects are so similar in function that their advantage lies almost exclusively in our aesthetic response. We choose objects that are likely to give us the most pleasure and thus make us happiest. So, next time you’re shopping for a new blouse, take comfort in the fact that, whatever you decide, your color preferences are at work helping you make the “right” choice. That’s their job! w
61
WAVEs of Color: An ecological valence theory of human color preference, an interview with Karen B. Schloss and Stephen E. Palmer is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.
1.
Ou, L.-C., Luo, M.R., Woodcock, A., & Wright, A. (2004a). A study of colour emotion and colour preference. Part I: Colour emotions for single colors. Color Research Application, 29, 232-240.
2.
Ou, L.-C., Luo, M.R., Woodcock, A., & Wright, A. (2004b). A study of colour emotion and colour preference. Part III: Colour preference modeling. Color Research Application, 29, 381-389.
3.
Hurlbert, A. C., Ling, Y. L. (2007). Biological components of sex differences in color preference. Current Biology 17, 623-625.
Photo Courtesy of Flickr MemberRyan Dickey
References
Playing (With) Color
Glimpse
www.glimpsejournal.com
by Fred Collopy, Case Western Reserve University
The necessity for the use of color in its various phases has only been felt by artists since 1800. The intensity of modern life has made a greater intensity necessary in art. Only by being more intense than life can art hold its own as a vital factor in either taste or inspiration.
I
— Stanton Macdonald-Wright, A Treatise on Color, 1924
didn’t know my grandfather; he died before I was born. But I did know a particular image of him. As a child, I used to visit a mural that was part of Weinold Reiss’ magnificent Cincin-
nati Union Terminal suite. There, my grandfather, who had posed for Reiss, stood at the
controls of a big green mixing machine with the most beautiful yellow liquid flowing into the 62
barrels below.
Figure 1. Detail from Weinold Reiss’ Cincinnati Union Terminal Mural. This panel was one of 14 mosaics, which were moved to the Northern Kentucky/Greater Cincinnati International Airport in the 1970s before Union Terminal’s concourse was destroyed. Image courtesy of Cincinnati Museum Center at Union Terminal, Cincinnati, Ohio.
And that’s what my grandfather witnessed each working day—the paint “as good as it was in the can.” Color, pure and simple. Color is an essential aspect of how we see the world. Whatever else visual artists set out to do, they cannot get around the need to attend to color. Even the absence of it constitutes a choice, since this, too, is a way of dealing with it. It is therefore no surprise that mankind’s fascination with visual instruments began with color. Over two centuries ago, dreamers started to imagine instruments that would use combinations and sequences of color to calm us, delight us, confound and intrigue us, much as musicians do with combinations and sequences of sounds. Color Scales
S
ir Isaac Newton puzzled over the nature of light and its relationship to sound. The physicist understood, as have many inventors and artists since, that both are wave phenomena that operate over a range of frequencies. With this knowledge, many pioneers became interested in creating an art of light—like the art of music—by creating instruments to make “color music.”2, 3 Newton associated each of the seven colors that he saw in prismatic light with the seven notes of the harmonic scale. Visual artist Karl Gerstner pointed out the randomness of this mapping, claiming that “Newton saw seven colors in the spectrum because he wanted to see seven colors, in order to correlate them with the notes and not vice versa...In actual fact observers can distinguish as many shades of color in the spectrum as they want: red, green, and blue
Still, Newton inspired a movement that has lasted for three centuries. In 1725, responding in part to Newton’s ideas about the nature of light, Louis Bertrand Castel, a Jesuit priest and physicist, wrote an essay announcing his invention of an ocular harpsichord. He proposed making colors transient the way that musical notes are. Using the keys of an ordinary harpsichord would reveal colors alone or in combinations, producing a succession that would elevate painting to the level of music. Castel proclaimed that his invention would lend colors “a certain vivacity and lightness which on an immobile and inanimate canvas they never have.”5 The idea of relating the notes of the musical scale to various colors occurred many times over the succeeding two centuries (Figure 2). Beyond their obvious arbitrariness, additional problems also exist with these scales.6 Perhaps the most significant is that they don’t account for how differently we make sense of visual and audio information. A pattern that sounds harmonious doesn’t necessarily look harmonious. In addition to relating various hues to particular musical notes, painters have explored the emotional content of colors and their potential for musical expression. Most famous among such characterizations can be found in Wassily Kandinsky’s classic essay, On the Spiritual in Art. Kandinsky describes each of the most common colors and compares them to musical sounds. He writes that “absolute green is the most peaceful color there is: it does not move in any direction, has no overtones of joy or sorrow or passion, demands nothing, calls out to no one...I would think the best way of char-
Color
I knew a wise-guy who used to make fun of my painting, but he didn’t like the Abstract Expressionists either. He said they would be good painters if they could only keep the paint as good as it is in the can. And that’s what I tried to do.1
being the most prominent, with fluid but short transitions.”4 Gerstner attributes this unscientific color mapping to Newton’s involvement in alchemy.
autumn 2009
Years after having visited the mural, as I got to know some of the work of modern artists, I was struck by a 1964 musing of Frank Stella, quoted in David Batchelor’s Chromophobia:
Sir Isaac Newton puzzled over the nature of light and its relationship to sound.
volume 2.3
As a paint-mixer for the Ault & Weiborg Corporation, he spent the long hours of each workday staring into large vats where he controlled levers and dials that transformed a few basic pigments into the variety of colors that the expanding demands of early 20th century taste required. Those who knew my grandfather said that he was a mellow man. Of course, inhaling paint fumes day after day could explain this reported temperament—though I like to think spending so much time with color also played a role.
63
www.glimpsejournal.com Glimpse 64
acterizing absolute green would be the quiet, expansive middle register of the violin.”7 The painter Stanton MacdonaldWright also characterized the emotional meanings of a dozen colors: “Yellow is superficial, has no depth of character, is frivolous, light, young-girlish and gay...Violet, we might sum up as being a cry, de profundis. It is the color of deepest depression, which comprises unhappiness, sorrow, silence, and the nearest approach to death in color.”8
therefore has something sad, an air of something sickly, something extinguished about it (like a slag heap!).”7
Kandinsky: “Absolute green would be the quiet,
But Kandinsky’s characterization of yellow is a different matter. He believed that “yellow, when directly observed (in some kind of geometrical form), is disquieting to the spectator, pricking him, stimulating him, revealing the nature of the power expressed in this color, which has an effect upon our sensibilities at once impudent and importunate.” Furthermore, he believed that yellow was “a color that inclines considerably toward the brighter tones, [that could] be raised to a pitch of intensity unbearable to the eye and to the spirit. Upon such intensification, it affects us like the shrill sound of a trumpet being played louder and louder, or the sound of a high-pitched fanfare.”7
expansive
middle register of the violin.”
Many of Kandinsky and Macdonald-Wright’s characterizations are in close agreement: Green is calm, quiet and peaceful; blue is ethereal, tranquil and heavenly; and red is deep, powerful and energetic. And Kandinsky’s description of violet evokes Macdonald-Wright’s: “Violet is thus a cooled-down red, in both a physical and a psychological sense. It
This depiction of yellow is hardly the “frivolous, light, younggirlish” thing we find in Macdonald-Wright’s characterization, and such differences of opinion underscore the challenge of establishing parallels between color and sound.
Figure 2. Three Centuries of Color Scales. Courtesy of the author. Sources: Newton: Isaac Newton, Opticks, 1704, Book 1, Part II, Proposition VI, Problem 2; Castel: Kenneth Peacock “Instruments to Perform Color-Music,” Leonardo (1988), 400; Field: Adrian Bernard Klein, Colour-Music, 1927, 69; Jameson: D. D. Jameson, Colour-Music, 1844, 12; Helmholz: Herman von Helmholz, Treatise on Physiological Optics, Vol. 2, 1962, 117; Bishop: Bainbridge Bishop, A Souvenir of the Color Organ, 1893, 11; Seeman: Klein, Colour-Music, 86; Rimington: Peacock, 402; Scriabin: Tom Douglas Jones, The Art of Light & Color, 1972, 104; Klein: Klein, 209; Aeppli: Gerstner, The Forms of Color, 169; Belmont: I.J. Belmont, The Modern Dilemma in Art, 1944, 226; Zieverink: Steve Zieverink, Twelve + Twelve, 2004, UnMuseum, Contemporary Arts Center, Cincinnati.
M
This notion of color context became prominent in the minds of color instrument designers. By the end of the 19th century, experimenters like Bainbridge Bishop turned their attention to how color music could evoke harmony and discord. Bishop believed that “by carefully examining the scale it will be seen that a direct contrast of color comes in as a discord—for example, a true green and red, or an orange and blue; but if we change the green for a bluish or yellowish green, the effect is much more harmonious.”10 Many theorists of the time—Ogden Rood and Percyval TudorHart among them—investigated the perceptual effects of various color combinations that would result in different visual harmonies and dissonances. These advancements in color theory inspired Stanton Macdonald-Wright and Morgan Russell, two of Tudor-Hart’s students, to found a new style of painting called Synchromism, a movement dedicated to the physical and emotional effects of color. Rood’s theory of harmonious
Macdonald-Wright’s response to the same basic impulse took a slightly different form. In 1919, he painted over five thousand pastels, each three by four feet. These were made into what he claimed was “the first full-length, stop-motion film ever made in color.”12 Russell and Macdonald-Wright saw their efforts as the creation of a new art—an art that differed from painting and dealt directly with time, movement and light itself. Stanton’s brother, Willard Huntington-Wright, drew on the experiences of the pair to speculate about the future of art. In a magnificent little book, The Future of Painting, he opens by arguing that “modernist painting” has little to do with painting as it has come to be understood. Instead, it is an accident that its practitioners have been driven to use pigments on canvas. Despite their struggles and the challenges of experimentation with new materials, what these artists are creat-
Color
Others, though, have taken advantage of the fact that colors are rarely perceived in isolation. They exist instead in a world of other colors and our perception of each color is affected by its context. In the early 19th century, Michel Eugène Chevreul, a French chemist and color theorist, studied this phenomenon. While working in a yarn-dyeing factory, he noticed that some colored yarn appeared different when seen in the context of other colors. He referred to this occurrence as “simultaneous contrast.”9
Russell and Macdonald-Wright both pushed their interest in color and light to the point where it lifted off of the canvas. For Russell, this took the form of a Kinetic Light Machine, which he first wrote about in 1912. He actually designed several machines, which were described in detail in his notebooks. A central feature of all of his machines was the use of rheostats, devices that allowed a “coming and dying out” effect with light. Russell wrote to Macdonald-Wright about his experiments: “I felt that abstraction led to lights and had fiddled with them in winter 1915-16 and late spring 1914 a bit.” And in a later letter he concluded: “As a matter of fact when I left off the synchromies in paint I got to meddling with lights also, but never got funds or encouragement...It was inevitable that synchromies should lead to this [manipulation of light].”11
autumn 2009
uch of the effect of music has to do with the simultaneous occurrence of multiple notes. The intervals between the notes create consonances and dissonances. The color theorist Wilhelm Ostwald (1921; cited in Gerstner4) proposed a system based on musical intervals. To each musical interval (e.g. the minor second) he assigned a shade composed of two color tones (e.g. a reddish yellow and a greenish yellow). By careful arrangement of his scale, he was able to capture many elements of musical consonance and dissonance.
color triads, or colors separated on the color wheel by 120 degrees, formed the bases of many of the Synchromists’ paintings. And Tudor-Hart’s conviction that sound and light are perceived in analogous ways became their inspiration.
volume 2.3
Color Harmonies
65
Glimpse
www.glimpsejournal.com
ing with color instruments is an art that is more of a performance, like a symphony, than a decorative art. The last chapter of Huntington-Wright’s essay opens this way:
66
The color-instrument of the future will not merely throw pretty squares, circles, coils, and volutes of colored light on a screen, but will be able to record the artist’s moods, desires and emotions along any visually formal aesthetic line. Only when such an instrument has been perfected can the modern artist’s creative conceptions be properly expressed. With the completion of this new medium the art of color will have entirely dissociated itself from the art of painting, not only in impulse and conception, but in the world’s attitude toward it.13
pretation of the intervals between color” and that “the closer an interval gets to the tritone [complementary color], the more interestingly/harmoniously it works as a color combination.”14 Color Tones
T
he discussions of scales and harmonies relate largely to how hues can affect emotion in a musical way. Controlling saturation and value can add expressive capabilities to a visual instrument in that the artist can achieve a variety of grays and tinted tones. An attention to saturation is illustrated in the work of one of the earliest color instrument designers, previously discussed.
In 1725...Louis Bertrand Castel, a Jesuit priest and physicist, proposed making colors transient the way that musical notes are.
Russell and Macdonald-Wrights’ color scale represents an important development in the art of “playing light.” Their scale revealed that soft and harmonious color combinations arise from pairing a color with one of its near-complements, or one of the colors on either side of its color complement. Likewise, they discovered that using complementary pairs would produce the color equivalent of dissonance in music. Macdonald-Wright instructed that “if a harsh clash is desired use red-orange and blue-green. For a clash less harsh use orange and blue. All sets of complementaries follow in the order of their harshness: yellow-orange and blue-violet; yellow-green and redviolet; green and red; and softest of all opposites, yellow and violet.”8 The intensity of the clash or dissonance is reduced as the colors are neutralized, by reducing their saturation or tonality (value).8 Taken together, these and other insights from the Synchromists provide a remarkably rich and musical approach to the construction of color harmonies and dissonances. Huntington-Wright sensed the importance of these insights for the developing art of light, and compared the use of color dissonance to the role that fortissimo passages play when we attend a symphony concert. Because they are such important musical concepts, current work continues to focus on harmony and dissonance. Katherine Lubar, for example, has explored color harmonies using Johannes Itten’s color wheel to create color intervals analogous to the intervals of tonal music. She went on to examine the color intervals for their similarity to the effects of corresponding musical intervals (consonance or dissonance). Among Lubar’s observations was that “tonal value plays an important role in the inter-
Bainbridge Bishop’s color organ placed a large white screen atop an ordinary organ. In addition to producing music, the controls pulled blinds down to expose stained glass windows of various colors through which light passed. In 1893, Bishop wrote about experiments in which he found that the use of tinted gray tones helped to produce a more musical quality to the light:
I soon found that a simple color did not give the sensation of a musical tone, but a color softened by gradations into neutral shades or tinted grays did so; also, that combinations of colors softened by gradations into neutral shades or tinted grays, with the edges of the main colors blending together, or nearly together, rendered the sensation of musical chords very well indeed.10 Additionally, changes in the value of a color alter how dark it appears, achieving an effect which can also have a fairly direct relationship to musical tone. Having done an extensive analysis of Van Gogh’s paintings and letters, Kurt Badt pointed to this relationship by noting that “dark and light colors do actually have effects which are comparable to low and high musical tones. Dark colors are sonorous, powerful, mighty like deep tones. But light colors, like those of the Impressionists, act, when they alone make up a whole work, with the magic of high voices: floating, light, youthful, carefree, and probably cool, too.”15
Concluding Remarks
C
OcuHarpsCastel.pdf) 6.
Collopy, Fred, “Color, Form and Motion: Dimensions of a Musical Art of Light,” Leonardo 33, No. 5, 355-360 (2000). (complete text available at http:// rhythmiclight.com/archives/bibliography.html).
7.
Kandinsky, Wassily, On the Spiritual in Art and Painting in Particular (Munich, 1912) In Kandinsky Complete Writings on Art, Lindsay, K.C. and Vergo, P. eds. (New York: DaCapo Press, 1994) 183, 189, 180-181.
8.
Macdonald-Wright, Stanton, A Treatise on Color, (Los Angeles, 1924) 19-20, 23, 23-24.
9.
Chevreul, Michel Eugene, The Principles of Harmony and Contrast of Colors, (1839), Excerpted in Primary Sources: Selected Writings on Color from Aristotle to Albers, Patricia Sloane, ed. (New York: Design Press,
10. Bishop, Bainbridge, A Souvenir of the Color Organ,
In my own experience, one does tend to land in places that are harmonious or dissonant when following their guidelines. Gradations do seem more musical and interesting, and darkness and lightness can effectively reinforce sonorous or light passages respectively. So, I continue to find inspiration in the work of these experimentalists and those who have followed them. Their efforts to articulate how various color elements work, and to reflect on and write about their choices, provide guidelines that serve us well in the absence of a comprehensive theory of color. w
12. Macdonald-Wright, Stanton, The Art of Stanton
with Some Suggestions in Regard to the Soul of the Rainbow and the Harmony of Light (New Russia, NY, 1893) 10-11, 4. (complete text available at http:// rhythmiclight.com/archives/index.html). 11. Kushner, Marilyn S., Morgan Russell (New York: Hudson Hills Press, 1990) 105. Macdonald Wright (Washington, D.C.: Smithsonian Press, 1967) 22. 13. Huntington-Wright, Willard, The Future of Painting (New York: B.W. Huebsch, Inc., 1923) 51. (complete text available at http://rhythmiclight.com/archives/ index.html). 14. Lubar, Katherine, “Color Intervals: Applying Concepts of Musical Consonance and Dissonance to Color,” Leonardo 37, 127-132 (2004). 15. Badt, Kurt, Die Farbenlehre van Gogh, (Cologne,
References
Germany: DuMont Schauberg, 1961), as cited in Gerstner (see 4).
1.
Batchelor, David, Chromophobia (London: Reaktion Book, 2000) 98.
2.
Klein, Adrian Bernard, Colour-Music: The Art of Light (London:
York: Design Press, 1989) 328. (complete text
Crosby Lockwood and Son, 1930).
available at http://rhythmiclight.com/archives/index.
Peacock, Kenneth, “Instruments to Perform Color-Music: Two
html).
3.
Centuries of Technological Experimentation,” Leonardo 21, No. 4, 397-406 (1988). (complete text available at http://rhythmiclight. com/archives/bibliography.html) 4.
Gerstner, Karl, The Forms of Color: The Interaction of Visual Elements (Cambridge, MA: MIT Press, 1986) 168, 173.
5.
Franssen, Maarten, “The Ocular Harpsichord of Louis-Bertrand Castel. The Science and Aesthetics of an Eighteenth-Century Cause Célèbre,” Tractrix: Yearbook for the History of Science, Medicine, Technology and Mathematics (1991) 15-77. (complete text available at www.tbm.tudelft.nl/live/pagina.jsp?id=1d0f8e1b-89e5-
Color
1991) 24-33.
Still, the Synchromists, Bainbridge Bishop and the others have managed to move us beyond the naïve mappings that dominated discussions of color music for so long. In wrestling as they have, at the level of perceptual issues, they have provided clues about how we should proceed in our consideration and advancement of this movement.
autumn 2009
apparent elegance. “Additive/subtractive theory had acquired a reputation for being scientific and technically unimpeachable, before it came to the attention of the editors of the Life science series. The question, as often in color theory, is how so ill-conceived an idea survived for so long.”16
41fa-8d4d-b90aed364d44&lang=en&binary=/doc/
volume 2.3
olor theory has not enjoyed the success that music theory has. Consider, for example, the widely disseminated theory of the additive and subtractive varieties of color mixing. Additive/ subtractive theory asserts that the rules by which colored lights combine are the inverse of those by which pigments combine. When the wavelengths on one color are added to another, they produce a color defined by the total. Similarly when one pigment is combined with another, it produces a color based upon the difference of their wavelengths. After describing some obvious problems with the theory, Patricia Sloane concludes that, as with much of color theory, the idea has little more going for it than its
16. Sloane, Patricia, The Visual Nature of Color (New
67
www.glimpsejournal.com Glimpse 68
Rel a ti vel y s p e ak i n g The relationship between language and thought in the color domain By Debi Roberson & J Richard Hanley
D
oes the “striped” appearance of the rainbow result from our habitual categorization of color appearance, or is it built into our perceptual systems by the innate organization of human color vision? If your language lacks separate
So much research has focused on color categorization because, excluding those with color vision deficits, humans can distinguish around 2 million nuances of color across the visible range. Alongside that commonality of visual experience, different languages use a widely varying number of color terms to describe their color sensations. Some languages use a single color term to refer to what an English speaker would describe as blue, green, or purple (Figure 1). Conversely, the range of colors that English speakers divide into two categories (green and blue) comprise three basic categories in Korean: chorok, cheongnok and parang.1 Even those languages with similar color vocabularies have slight variations in the range of stimuli covered by a particular term (e.g. the different ranges covered by the English term blue and the Italian term blu).
Humans can distinguish around
2 million nuances of color
across the visible range...Different languages use...varying...color terms to describe their color sensations terms for “blue” and “green” (as many languages do, including Welsh) would you perceive these shades as more similar than a speaker of English? Although the nature of the relationship between natural language and our mental representation of the experienced world has been probed by philosophers, psychologists, linguists and anthropologists in many areas, the color domain has long been in the spotlight of investigation.
This strikingly diverse linguistic categorization of an apparently common perceptual experience has led to a prolonged debate about the nature of the relationship between sensation, cognition
Figure 1. Distribution of Himba named categories and choices of best exemplar for the 160 chip saturated array (for 31 observers) compared to those of English and Berinmo speakers for the same array. Numbers represent number of individuals choosing an exemplar as best example of the category. Dots on English graph represent the locations of best examples for English speakers Adapted from Roberson, D., Davidoff, J., Davies, I. and Shapiro, L., “Colour categories in Himba: Evidence for the cultural relativity hypothesis,” Cognitive Psychology 50, 378-411 (2005).
Himba Color Naming Distribution 5R
10R
5YR 10YR 5Y
vapa
9
1
1
1
dumbu
1
3
4
4
3
serandu
burou
1 2
2
2
1
1
1
2
1
1
1
1
1
5AP 10RP 2
2
1
serandu 1
1
1
1
1 1
1 1
2
1
1
2
vapa
2
10P
5 1
1
zoozu
1
1
2
1
5R 9
10R 1
1
5YR 10YR 5Y
Wap
8 1
7
Mehi
6 5
3
4
10
1
4
4
3
3
2
2
Wap
1
1
2
1 1
5
8
4
2
1
1
4
Noi
1 2
3
7
Mehi
2
2
1 1
2
1
5AP 10RP
1
2
Kei 1
1
10P
1
5
1
10B 5PB 10PB 5P
2
Wor
3
1
2
5GY 10GY 5G 10G 5BG 10BG 5B
1
1
1
3
10Y
Color
Berinmo Color Naming Distribution
2
5
Kei
1 1
1
1
4
1 2
2
English Color Naming Distribution 5R
9 8 7 6 5 4 3 2
.
10R
PINK
5YR 10YR 5Y
.
.
RED
5GY 10GY 5G 10G 5BG 10BG 5B
10B 5PB 10PB 5P
10P
5AP 10RP
YELLOW
ORANGE
.
10Y
.
BROWN
. GREEN
PINK
.
BLUE
autumn 2009
3
5
10B 5PB 10PB 5P
volume 2.3
6
2
1 1
2
1
5
1 2
2
7
5GY 10GY 5G 10G 5BG 10BG 5B
2
1
8
10Y
.
PURPLE RED
69
www.glimpsejournal.com
and language. At one extreme of the debate has been the view that all humans share both perceptual experience of color and mental organization of that experience into universal “basic” categories.2, 3 Under this account, different languages label those experiences differently because some color vocabularies are still evolving towards the ideal optimal arrangement (represented by the set that are named in English – red, yellow, blue, green, orange, purple, pink and brown, together with black, white and grey). In contrast, other researchers have suggested that only the level of raw color sensation is universal; categorical divisions, and their labels, are learned and vary according to the cultural needs of different societies.4, 5 ,6, 7, 8, 9
Glimpse
Such wide variability of color naming across different languages would be consistent with the proposal that learning a particular language influences which categorical distinctions one develops.4, 10 This would not be true, however, if the categories that we name (e.g. red or green) can be functionally independent of our mental organization of color sensations. 11, 12, 13 In that case we would have one set of categories in mind, but a completely different system in our language.
70
Heider and Olivier reported that Dani observers in Irian Jaya showed a similar pattern of memory for colors to that of English speakers, but would choose from just two categorical labels to name all the stimuli. 3 The authors interpreted this as evidence that these observers had one set of categories in mind, but an orthogonal set that they expressed when naming colors. A complete disconnection of that nature between thought and language would have widespread theoretical implications for many areas of cognition. For English speakers, such a disconnection seems unlikely, because colors that are easier to name in English are easier for English speakers to remember accurately and easier to communicate to others.3, 14, 15, 16 A similar relationship between naming and memory has now been shown for Spanish,10, 17 Berinmo (a Papuan language)18 and Himba9. This evidence suggests that ease-of-naming might be a good predictor of memory accuracy in any language regardless of whether it has four color terms or twenty. But what if some ways of organizing colors into categories are better than others? What if Western languages such as English
(a) (b)
B
“green”
A
D
“blue”
C
have reached the endpoint of an evolutionary trajectory and use an optimal set of color terms and categories? Since all humans would share underlying cognitive representations of the optimal organization, color vocabularies that do not express all the optimal categories in their language would eventually evolve towards the optimal set. This organization at a language-independent cognitive level might be innate and hardwired into the organization of color vision pathways.13, 19
autumn 2009 Color
Recent investigations have sought to overcome these limitations by using tasks without a memory compo-
volume 2.3
Under such a view, no learning or transmission of cognitive color categories would be required (since everyone would have the same set from birth). However, speakers of languages that express a different (non-optimal) number of categories would still at some point have to learn the appropriate reference set of exemplars for the linguistic terms that were appropriate in their culture.13 It remains unclear how a detailed category structure of this nature could be innately hardwired, given both the general flexibility of developing cognitive systems and the ease with which humans can learn new color categories.20 It is also unclear why or how differences between linguistic and cognitive categorization come to exist in some cultures, but not in others.
Empirical and conceptual difficulties exist for both positions outlined above. On the one hand, early experiments used arrays of colors for testing that were not equated for difficulty because the array from which participants were required to recognize target colors was ordered by hue and brightness. This inflated recognition rates for best example red and yellow, which appeared at the edges of the array, and thus had fewer neighboring stimuli than any intermediate stimuli. Randomization of the array18 removed the apparent recognition advantage for central exemplars of the proposed “universal� categories. In addition to possible artifacts resulting from the use of particular stimulus arrays, the use of memory tasks to probe the cognitive organization of color categories may have encouraged participants to rely strategically on verbal labels during the retention interval, creating a falsely inflated impression of the relationship between language and thought.21
71
( )
( )
Figure 2. The visual search task employed by Gilbert et al. (2006) (a) Print-rendered versions of the four colors used. (b) Sample display for the visual search task. Participants were required to press one of two response keys, indicating the side containing the target color. (c) In the no-interference condition, RTs were faster for the between-category pair and slower for the within-category pairs when targets appeared in the RVF compared with when they appeared in the LVF. (d) Effects were reversed with verbal interference.
Adapted from PNAS January 10, 2006 vol. 103 no. 2 489-494
(b)
2.5 B
1.25 B
10 BG
8.25 BG
7.5 BG
7.5 B
40 5B
3.75 B
2.5 B
1.25 B
10 BG
8.25 BG
7.5 BG
7.5 B
6.25 B
Munsell Hue
6.25 B
5 BG
3.75 BG
2.5 BG
% Named
www.glimpsejournal.com
60
5B
3.75 B
Munsell Hue 6.25 BG
72
6.25 BG
5 BG
3.75 BG
2.5 BG
1.25 BG
10 G
(a)
1.25 BG
10 G
% Named
Glimpse  
100
80
Blue Green
40
20
0
100
80
60
Chorok Cheongnok Parang
20
0
100 Korean English
60 volume 2.3
40
autumn 2009
20
nent. The most fruitful of these investigations have involved categorical perception (CP) of color in speakers of a range of languages. CP is said to occur when a continuous change along a perceptual dimension comes to be judged as a series of discrete qualitative regions separated by sharp boundaries between labeled categories.22 Consequently, two adjacent items on a continuum are more easily judged as different when they come from different categories (i.e. one blue and one green) than two equally separated items from the same category (i.e. both blue). Gilbert et al. devised a matching-to-sample visualsearch task that appeared to make little or no demands on memory to investigate CP for colored targets.23 Participants were told to fixate on a cross in the center of a computer screen. They then reported the location of an “oddball” colored target appearing amongst an array of distractors that were identically colored (Figure 2). Participants showed clear evidence of CP on this task. That is, they were faster to detect a difference between the target and background when the target and background colors came from different categories (e.g. blue target, green background) than when both target and background came from the same category
7.5 B
6.25 B
5B
3.75 B
2.5 B
1.25 B
10 BG
8.25 BG
7.5 BG
6.25 BG
5 BG
3.75 BG
2.5 BG
1.25 BG
(c)
Color
0 10 G
∆ Threshold
80
Munsell Hue (e.g. different shades of blue) even when the amount of physical separation between target and background were held constant. A critical question is whether CP on this visual-search task occurs only at boundaries between colors in the putative universal set or whether it also occurs at boundaries that are not marked in English. This issue has been recently investigated with speakers of Russian and with speakers of Korean. Winawer et al’s Russian participants showed CP at the boundary between siniy (dark blue) and goluboy (light blue), which are distinct “basic” color terms for speakers of Russian. 24 English speakers, who would call all these stimuli “blue”, did not show the same cross-category advantage. Roberson et al’s Korean participants showed CP at the boundary between yeondu (yellowgreen) and chorok (green), which are distinct “basic” color terms for speakers of Korean but not for speakers of English. 25 Native English speakers did not show evidence of CP for either boundary.
Figure 3. Mean naming (a) for English speakers and (b) for Korean speakers across the range of stimuli tested from the centre of the green category to the centre of the blue category and (c) JND discrimination thresholds (in ∆E units) for English and Korean speakers across the same range of stimuli. Adapted from Roberson, D., Hanley, J.R., & Pak, H.S. (2009) Thresholds for colour discrimination in English and Korean speakers. Cognition, 112, 482-487.
73
www.glimpsejournal.com Glimpse   74
Because the experimental tasks were not tests of memory, these two studies provide a clear demonstration that culture and language constrain a categorical perception of color. Consequently, these two studies provide overwhelming evidence that superior discrimination of stimuli that cross a category boundary (such as that found for English speakers at the boundary between blue and green) cannot be considered sufficient evidence for a set of universal color categories that are hard-wired in the human visual system. CP effects were once thought to reflect low-level stages of early visual processing.26 Sensory mechanisms might be permanently re-tuned to be more sensitive to changes at the boundary than in the category centers. A recent experiment with both English and Korean speakers found that both populations showed CP in same-different judgments for stimuli ranging across the blue-green boundary that they could easily tell apart. Despite this, for speakers of both languages, discrimination thresholds (the point at which they could no longer reliably distinguish two shades of color) did
losum, which connects the two hemispheres of the brain, had been surgically severed.23 A recent functional Magnetic Resonance Imaging (fMRI) study provided further evidence that left-hemisphere brain regions associated with language processing are actively associated with early perceptual processing of color.28 Easy-to-name colors evoked stronger activation in areas associated with language than hard-to-name colors. The authors suggest that these results support the rapid automatic activation of verbal color codes during perceptual decisions about color. VanRullen and Thorpe29 have suggested that stimulus identification and categorization should be thought of as concurrent, rather than sequential processes, and early effects of language on brain responses support that view.30, 31 It is not hard to produce an explanation of the way in which lefthemisphere language processes might produce categorical perception. Let us assume that decisions about whether a target and background are the same can be made on the basis of either a right hemisphere perceptual code or a left hemisphere verbal code, and that when the two codes conflict, accuracy and speed will be reduced. Automatic activation of color names should therefore impair judgments about whether, for example, two different shades of blue are from the same category because the linguistic information that they are the same will conflict with the perceptual information that they are different. Decisions for items from different categories (e.g. a blue and a green) will be faster and more accurate because both linguistic and perceptual codes indicate that target and background are different. When the left-hemisphere language system is suppressed by verbal interference, or is not accessed because information is presented directly to the right hemisphere, the verbal code is not generated and there is never any source of conflict with the perceptual code. Hence there is no advantage for comparisons that fall across linguistic category boundaries.
Does linguistic processing affect the early stages of color processing? not differ across a range including both good examples of blue and green and stimuli that lay on the borderline between those categories (Figure 3). 27 Is it therefore the case that linguistic processing affects the early stages of color processing? Two sets of additional findings have provided information about the precise point at which verbal codes influence color categorization. First, Gilbert et al.23 and Winawer et al.24 showed that participants did not demonstrate CP in cases where they also carried out a concurrent verbal task. Under verbal suppression, all equally-spaced separations of color were equally easy to discriminate. Second, Gilbert et al.23 and Roberson et al.25 reported that the CP effect was only found for colors that were presented in the right visual field, presumed to preferentially access language-processing areas in the left hemisphere. No difference between within- and acrosscategory pairings of targets and distractors was observed for colors presented in the left visual field, which gains preferential access to the right hemisphere. Gilbert et al. also showed that CP was found only in the left hemisphere of a patient in whom the corpus cal-
If this account of categorical perception is true, then it follows that the ability to decide whether two shades of color are different probably depends on perceptual processing systems located either in the right hemisphere or bilaterally, and does not require any form of verbal mediation. Color categorization however, is entirely the product of left hemisphere categorization systems closely linked to language processing. Because both verbal and perceptual codes for color are automatically activated relatively rapidly, conflicting codes can produce errors. But where these two sources of information yield congruent information, decisions about color can be made accurately and rapidly, and memory for colored stimuli is good. It therefore appears that categorical perception of color, contrary to what has often been claimed, does not in fact reflect superior discrimination of colors when they
cross a category boundary. Instead it appears to reflect the fact that decisions about color are hampered when perceptual codes and verbal codes are in conflict. This conflict occurs when a task requires that two different shades of a primary color must be treated as different even though they share the same label.
decisions about
color are hampered
codes and verbal
codes are in
conflict.
Several other outstanding questions remain that are both fundamental to the debate and beyond the scope of empirical investigations to date. Is the development of adult color categorization a unique case? If not, to what extent does it follow a similar pattern to other modalities that come to be perceived categorically? In studies of object classification, evidence from cross-linguistic studies best fits a hybrid model in which some broad, shared, nonlinguistic understanding of a domain combines with varying cultural pressures to differentiate particular aspects of a dimension at particular times in
There does appear to be a separate non-linguistic system (possibly in the right hemisphere) that can make extremely fine discriminations between shades of color. There is no evidence that language learning has any effect on the way that this system processes color. We do not believe, however, that this system “knows” precise information concerning similarities and differences between two shades of color (e.g. that one is brighter than another, that one is more saturated than another, or that two different shades may share the same name). We do not believe, therefore, that this is the system that makes us see the rainbow as comprising seven distinct colors. Categorical knowledge of this kind is only available to the left hemisphere language-based color system, and people with different linguistic categories may well see a smaller or greater number of colors in the rainbow as a consequence. w
References 1.
Kim, Y. S., Pak, H. and Lee, Y. H., “A study on Munsell color space for Korean color names ( I ),” Journal of Korean Society of Color Studies 15, 29-36 (2001).
2.
Berlin, B. and Kay, P., Basic color terms: Their universality and evolution, (Berkeley: University of California Press, 1969).
Color
when perceptual
autumn 2009
Early research in the field led to the conclusion that there are separate levels of categorical representation of color, one cognitive and impervious to language and another more superficial linguistic level. 3, 11 The evidence reviewed here also points to the existence of distinct linguistic and non-linguistic color systems. However, there is no evidence that the linguistic system is in any way superficial. Linguistic categorization in different languages and cultures partitions the same range of visible colors in different ways and these differences affect decisions about color even on visual search tasks. Evidence suggests that categorical effects in color perception and memory occur as a result of access to lexical codes for color.
volume 2.3
Although most researchers now agree that categorical perception of color appears to be highly variable across cultures and languagedependent, many outstanding questions remain to be addressed with regard to the different ways that linguistic communities divide the continuum of color. Observers can rapidly learn new color category divisions to the point that they show CP for the newly-learned boundary.20 So, if an eventual set of eleven basic categories were optimal, why do some cultures maintain a small set of linguistic categories while surrounded by other languages that have larger sets? Why would languages like Russian, Greek, Italian, Turkish, Vietnamese and Korean develop additional basic color terms beyond those used in English? Whatever the origin of the observed differences between the color terminologies of different societies, any comprehensive model of color categorization needs to explain both the observed similarities and the differences between color naming systems. The origins of linguistic color categories in different societies might be constrained by either cultural or environmental needs, or both, and both may change, over time, in different ways in different communities.
their history.32 Recent computational models of color category instantiation support the view that a combination of shared domain structure (in terms of both available learning mechanisms and the range of visible colors in the environment) and language is needed to explain shared color category structure.33, 34 Such a combination of factors might lead to differences between linguistic categorization systems that also vary depending on the degree (and nature) of interactions between linguistic communities.
75
3.
4.
Heider, E.R. and Olivier, D.C., “The structure of the color
perception: A perceptual learning approach to the linguistic
Cognitive Psychology 3, 337-354 (1972).
relativity hypothesis,” Journal of Experimental Psychology: General
Ratner, C., “A sociohistorical critique of naturalistic theories of color perception,” Journal of Mind and
5.
www.glimpsejournal.com Glimpse
7.
spatial representation: Setting some boundaries,” in Language in
Gumperz, J.J. and Levinson, S.C., “Rethinking linguistic
mind: Advances in the study of language and thought, Gentner, D. and
Kuschel, R. and Monberg, T., “‘We don’t talk much about
perception: A critical overview,” in Categorical perception: The
Island,” Man 9, 213-242 (1974).
groundwork of cognition, S. Harnad, ed., (Cambridge: Cambridge
Davidoff, J., Davies, I. and Roberson, D., “Colour and 402, 604 (1999). Roberson, D., Davidoff, J., Davies, I.R.L. and Shapiro, L. R., “The Development of Color categories in two
9.
University Press, 1987). 23. Gilbert, A.L., Regier, T., Kay, P. and Ivry, R.B., “Whorf is supported in the right visual field but not the left,” Proceedings of the National Academy of Sciences 103, 489-494 (2006). 24. Winawer, J., Witthoft, N., Frank, M.C., Wu, L. and Boroditsky, L.,
languages: A longitudinal study,” Journal of Experimental
“Russian blues reveal effects of language on color discrimination,”
Psychology: General 133, 554-571 (2004).
Proceedings of the National Academy of Sciences 104, 7780-7785
Roberson, D., Davidoff, J., Davies, I.R.L. and Shapiro, L., “Colour categories in Himba: Evidence for the cultural relativity hypothesis,” Cognitive Psychology 50, 378-411
76
Goldin-Meadow, S. eds. (Cambridge: MA: MIT Press, (2003). 22. Harnad, S., “Psychophysical and cognitive aspects of categorical
colour here’: A study of colour semantics on Bellona
categories of a stone-age tribe,” Nature 398, 203-204 8.
131, 477–493 (2002). 21. Munnich, E. and Landau, B., “The effects of spatial language on
Behavior 10, 361-373 (1989). relativity,” Current Anthropology 32, 613-623 (1997). 6.
20. Özgen, E. and Davies, I.R.L., “Acquisition of categorical color
space in naming and memory for two languages,”
(2005). 10. Lantz, D. and Stefflre, V., “Language and cognition
(2007). 25. Roberson, D., Pak, H. and Hanley, J.R., “Categorical perception of colour in the left and right hemisphere is verbally mediated: Evidence from Korean,” Cognition 107, 752-762 (2008). 26. Bornstein, M.H. and Korda, N.O., “Discrimination and matching within
revisited,” Journal of Abnormal and Social Psychology
and between hues measured by reaction times: Some implications
69, 472-481 (1964).
for categorical perception and levels of information processing,”
11. Rosch Heider, E., “Universals in color naming and memory,” Journal of Experimental Psychology 93, 10-20 (1972a). 12. Rosch Heider, E., “Natural categories,” Cognitive Psychology 4, 328-350 (1973). 13. Bornstein, M.H., “On the development of color naming in young children: Data and theory,” Brain and Language 26, 72-93 (1985). 14. Brown, R. and Lenneberg, E., “A study in language and cognition,” Journal of Abnormal and Social Psychology 49, 454-462 (1954). 15. Garro, L.C., “Language, memory and focality: A
Psychological Research 46, 207-222 (1984). 27. Roberson, D., Hanley, J.R., & Pak, H.S. (2009) Thresholds for colour discrimination in English and Korean speakers. Cognition, 112, 482-487 (2009). 28. Tan, L.H., Chan, A.H.D., Kay, P., Khong, P-L., Yip, L.K.C. and Luke, K-K, “Language affects patterns of brain activation associated with perceptual decision,” Proceedings of the National Academy of Sciences 105, 4004-4009 (2008). 29. VanRullen, R. and Thorpe, S.J., “The time course of visual processing: from early perception to decision-making,” Journal of Cognitive Neuroscience 13, 454-461 (2001). 30. Thierry, G., Athanasopoulos, P., Wiggett, A., Dering, B. and Kuipers,
re-examination,” American Anthropologist 88, 128-136
J-R., “Unconscious effects of language-specific terminology on
(1986).
preattentive color perception,” Proceedings of the National Academy
16. Agrillo, C. and Roberson, D., “Colour language and colour cognition: Brown & Lenneberg revisited,” Visual Cognition 17, 412-430 (2009). 17. Stefflre, V., Castillo Vales, V. and Morley, L., “Language and cognition in Yucatan: A cross-cultural replication,” Journal of Personality and Social Psychology 4, 112-115 (1966). 18. Roberson, D., Davies I. and Davidoff, J., “Color categories are not universal: Replications and new evidence from a Stone-Age Culture,” Journal of Experimental Psychology General 129, 369-98 (2000). 19. Kay, P. and McDaniel, C.K., “The linguistic significance
of Sciences of the USA 106, 4567-4570 (2009). 31. Siok, W.T., Kay, P., Wang, W.S.Y., Chan, A.H.D., Chen, L., KangKwong, L. and Tan, L-H., “Language regions of brain are operative in color perception,” Proceedings of the National Academy of Sciences of the USA. 106, 8140-8146 (2009). 32. Malt, B.C., Sloman, S.A. and Gennari, S.P., “Universality and language specificity in object naming,” Journal of Memory and Language 49, 20-42 (2003). 33. Steels, L. and Belpaeme, T., “Coordinating perceptually grounded categories through language: A case study for colour,” Behavioral and Brain Sciences 28, 469-489 (2005). 34. Belpaeme, T. and Bleys, J., “Explaining universal color categories
of the meanings of basic colour terms,” Language 54
through a constrained acquisition process,” Adaptive Behavior 13,
(3), 610-646 (1978).
293-310 (2005).
volume 2.3
autumn 2009 Color
77
Color-Struck Quilting and Colorism in the African-American Community
Glimpse
www.glimpsejournal.com
by Lauren Cross When I look back at my childhood, I can recall a significant link between memory and color. From the lemon-yellow purse that my mother gave me as child to the ever so popular peach bridesmaid dresses of the 1980s, the variations of color filled my vision and influenced my perception of humanity. Like most young children, I was embracing the connection between the colors and images I was attracted to, and opening my understanding to the meanings behind them. Yet, I do recall a pivotal moment while entering grade school where my impression of the hidden power of color began to change. It was a time that I’ve had a hard time negating from my memory. I was at the tender age where children begin to learn how to interact with one another--a time where we began to notice our differences.
78
a childhood friend... insisted that because I was of a lighter complexion, I was not “black,” but “tan.” While in primary school, I experienced my first case of skin color prejudice, yet surprisingly not from a fellow white classmate (which I felt my parents prepared me for), but from a person of my own African-American race. I began to notice distinct references made to my skin color from other African-American classmates and friends that I had never heard before. I was called names like “yellow bone” (a light-skinned African-American woman) or “red” (used to describe an African-American with Native-American heritage) because of the length of my long hair (Chris Rock’s new Good Hair documentary talks about this). Then, a monumental event occurred when I got into a terrible argument with a childhood friend, who insisted that because I was of a lighter complexion, I was not “black,” but “tan.” This confused me, as my parents had always encouraged me to appreciate my black heritage. Yet, my friend was convinced that because of my skin color, I had none at all.
From that point on I began to realize that color has a meaning outside of the visual aesthetics of objects and images that create our culture. Color, when integrated into the social context of our society, is influenced by the politics of class and race. Racism, which to my belief is the origin of most prejudice of human skin color, seemed to be contributing to the politics of skin color prejudice within individual minority groups. It seemed that the spread of attitudes of “white supremacy” created a subconscious prejudice that trickled into the minds of many. Everyone wanted to be pure; everyone wanted to be “white”. The color “black” seemed to be deemed lowest in value on the racial scale--black also carrying the connotation of all things evil and bad. Though the world often thinks in black and white, the reality is that there are many ranges of tones that exist in the black community: from olive, to cinnamon, to dark brown. Though the AfricanAmerican community once declared that “Black is beautiful,” it seems that some really don’t believe it. Because “near-white” skin has always seemed to present more
What did it mean? volume 2.3 autumn 2009
opportunities for African-Americans of lighter complexion (when amongst white-led groups, organizations and employers), blacks in turn perceive that having light skin is better for surviving racism. We can see aspects of this in Kenneth and Mamie Clark’s “doll test” that was used to show the effects of segregation in Brown vs. Board of Education. Young black children between the ages of three and seven were asked to choose which doll (black or white) they preferred. The majority of the children chose the white doll. We again witnessed this disturbing reality fifty years after Brown vs. Board of Education in the documentary A Girl Like Me, by Kiri Davis (2005). In addition to interviewing young black women who express their personal issues with skin color in the AfricanAmerican community, Davis’ seven-minute film recreates the Clarks’ experiment with twenty-one young black children. The majority of children in the film, like those in Clark’s study, preferred to play with the white doll and considered it “nice” compared to the black doll, which they considered “bad.”
Color
Coming from a family of all shades, the notion of not appreciating a person because of the hue or shade of their skin was something I was never brought up to believe. I later learned that this was a different experience for many of my classmates. In comparison, their awareness of skin color was ingrained, or as some might say, they were “color struck.” 79
What was this fascination with human color? When I was in grade school, I observed the effects of this mentality with horror as my fellow “black” classmates ridiculed African students because of the “authenticity” of their African roots. It was clear that my classmates were, for some reason, ashamed of their African heritage. At the time, I questioned where these ideas came from. I began to associate my classmates’ thoughts with the influence of the 1980s media, where the representation of the African image at that time was both demeaning and unjust. I recall awkward moments in U.S. History class where some students would find humor in the images of darker-skinned, “naked Africans” that were shown in videos or films. They went so far as to make derogatory references to animals, such as monkeys or gorillas. I found this to conflict with what I knew were the same stereotypical ideas that were projected upon Africans during slavery. How does an African-American who has heritage in Africa laugh at their own image? Evidently, my classmates had internalized these messages. Over the years, questions filled my mind. Where was the “black pride” that my father spoke of? Did my classmates’ parents not inform them? What was this fascination with human color? What did it mean?
Although I was unaware at the time, this was a major, yet secretive issue in the black community, specifically in the American South, where the roots of racism still run deep. Much later in my adult life, I learned how to call this issue by name: colorism. Though the term colorism is a fairly new word, it’s undeniable amongst black people, once explained. Whenever I describe the act of skin color prejudice among family and friends, they automatically know what I mean, whether they’ve experienced it personally or not. What surprised me most was that skin color politics was an issue not only prevalent in the black community, but an issue also experienced in Latino, Asian and Indian cultures. This is illustrated in the advertising for bleaching creams that target
(their) awareness of skin color was ingrained, or as some might say, they were “color struck.”
Glimpse
www.glimpsejournal.com
women in minority groups, as well as in the writings of scholars and writers like Joanne Rondilla (Is Lighter Better? Skin Tone Discrimination among Asian Americans) and Margaret Hunter (Race, Gender, and the Politics of Skin Tone), who have written about the social and economic advantages of having a lighter complexion. I did a significant amount of research in graduate school on the subject of colorism because I wanted to wrap my head around the issue and to start creating artwork about the idea of color defining our human connection to social privilege. I read several books ranging from autobiographical accounts to research that allowed me to see the dynamics of the issue and to discover that other scholars were investigating these issues as well. Growing up, I heard much bitter talk about the effects of colorism from friends and classmates who were much darker than me. Being an African-American woman of caramel-light skin tone, I often felt silenced when talking about color because I was neither “dark enough,” nor “light enough” to complain. I was accused of being from
I wanted to wrap my head around the issue ...of color defining our human connection to social privilege. also received preferential treatment by other black colleagues because of my skin color, or have been patronized for my achievements whenever it appears that I’ve received a higher level of recognition than they have. In their eyes, my success is only partial. Though I wish I could say this was untrue, it’s difficult when studies have shown that when choosing between a light-skinned and dark-skinned employee, most employers pick the lighter.
80
In addition to the economic and social issues that arise within the African American community as a result of colorism, the personal inter-relationships can also be strained. I can’t possibly name the amount of times I’ve been approached by black
a background of house slaves and mulattoes, or living a sheltered life. “What do you have to complain about?” Some even called me white. I understood their argument, yet regardless of how false or true their assumptions were, it never feels good to be the outcast. It seemed that no matter how hard I tried to express my heart about the issue, I was always accused of not being worthy of discussing the issue because of the advantages of lighter skin. I have to admit, there have been a number of instances where I’ve felt I was offered a particular job because of the color of my skin in addtion to the professionalism that I exhibit. I’ve also seen this preference occur in how I’m treated in business situations compared to other colleagues who are darker than me. In one job, while I was treated with the utmost respect for the work that I did, a fellow employee of darker skin tone was treated more harshly, as if they’d achieved nothing at all. Yet, I’ve
males who were attracted to me because they preferred lightskinned women with more European features; they believed that dating such a woman would help them “move ahead” in society. I was even told by one man that light-skinned or yellow-colored women were the best alternative to dating outside of their race. This also caused an issue among my female friends because of the “competitiveness” of trying to gain the attention of men.
I became interested in this phenomenon, and wanted to determine whether quilting provided an avenue for African American women and men to begin to appreciate the “skin they’re in.” So, I began a documentary film project called The Skin Quilt Project (www. skinquiltproject.com), and through its production, I’ve found that quilting is an art form that provides an intimate connection with cultural history, tradition and the opportunity to gain fellowship, self-esteem and self-confidence. Quilters, as one participant described, have the deepest appreciation for color; through their
Color
ery, where it has lifted people of African descent into a place of value and cultural treasure. By combining pieces of fabric, ranging from the wedding dresses to the uniforms of ancestors, quilts essentially serve as one of the major records of African American culture from slavery until today.
autumn 2009
Over the past two years, I’ve learned how significant the quilting tradition has been in the African-American community since slav-
volume 2.3
All of these experiences and observations encouraged me to find individuals in the African-American community who had suffered from the effects of colorism, yet found a way to overcome the issue, while learning to appreciate their African heritage. Over time, I observed this pride and appreciation of heritage among AfricanAmerican quilters. At different times in my life, I observed several family friends begin their journey into quilting, and I watched in awe as their confidence unfolded in the richest of ways. They had received a life-changing gift that could never be taken away.
love of color, they have a greater appreciation for the beauty in themselves and others. Yet, for those who have dealt with issues of colorism in the past, they learn to overcome a horrible mentality passed down from slavery, making them believe 81 that they would never succeed in anything, based on their skin color alone. Through quilting, however, these individuals experience healing through developing their craft and using their own creative skill. In speaking with a number of quilters, my own perception of the African-American heritage has been enriched; after all, both my great-grandmother and grandmother were quilters. And though the media may influence our culture to believe that black beauty is defined by light skin, I hold to my strong conviction that beauty truly is in the “eyes of the beholder.” I’m motivated to increase the awareness of colorism in the African- American community so that the distortion of the black image will end. Being creative and engaging with the positive impacts of one’s cultural history can help us all appreciate what it means to be individuals. I’ve started my own quest, by picking up fabric, needle and thread, and letting the journey begin. w Visit the www.glimpsejournal.com blog or www.skinquiltproject. com for a video trailer of The Skin Quilt Project.
Human Potential for Tetrachromacy www.glimpsejournal.com
by Kimberly A. Jameson
Glimpse
Humanly Visible Region
Lower
Higher
82
[waves per second]
Figure 1. The Sun’s electromagnetic spectrum with the small portion that is visible to humans highlighted by pseudo-coloring. Scale shown is approximate, created based on information from The U.S. Department of Energy Lawrence Berkeley National Laboratory website (www.lbl.gov/MicroWorlds/ALSTool/EMSpec/EMSpec2.html).
N
ature’s color palette—the changing sky, autumn leaves, the tinted irises of beloved eyes—has allured human interest since time immemorial. Scientific advances over the past twenty years have led to a far better understanding of the relevance and physiological basis of color experience than ever before. Recent research in molecular genetics, color perception and cognitive psychology is clarifying the underpinnings of human color sensations, how color experience has evolved, and along which perceptual paths we might be headed as a species of color-experiencing individuals. Together, such advances suggest that extensions of color perception theory are needed to account for retinal photopigment diversity unanticipated by accepted models of color vision trichromacy. Why do we experience color?
T
he ability to perceive color is so natural that we rarely consider its origin. Color perception, like
perception of texture or motion, occurs when our visual system encounters illuminated objects. This ability to detect surface variation by sampling the light, or spectra, reflected off environmental objects is widespread across species. Humans enjoy color by processing reflected spectra within a narrow (~380 nm to ~780 nm) “visible” range of electromagnetic wavelengths (Figure 1). Color requires both (i) photon capture by photoreceptors and (ii) encoding of photoreceptor excitation ratios. The number of colors humans can distinguish varies across individuals, and is generally estimated to be between one and ten million. Perceived color variation is due to the ways our available photoreceptors react to reflected light. Photoreceptor response sensitivities also underlie metameric color equivalence classes (object reflectance spectra that have different physical forms but produce the same color percept) (Figure 2). The existence of natural and man-made metameric
What is color vision for?
D
Humans derive color information from responses of typically three cone classes containing different photopigments, distributed by the millions across the retina. These different cone classes are generated through expression of different opsin genes. Opsin genes with different amino acid sequences and a light-absorbing chromophore can produce photoreceptor classes with drastically different absorption spectra.
Color 83
Genetic sequences identified for human light-sensitive pigments include: (a) the chromosome-3 linked rod rhodopsin pigment9 that interacts with color vision at low
Munsell 5P 6/8 approximated
Reflectance Intensity
espite the non-unique mapping from color to reflected light, color cues are used in detecting targets against dappled backgrounds, perceptual segregation and object identification or categorization by color.1 During non-human primate evolution, an ability to detect color differences from surface reflectances was likely selected for because it provided a means of signaling for the species. Perhaps color permitted the identification of carbohydrate rich fruit or tender leaves,1, 2, 3, 4, 5 or aided social interaction through detecting physiological states of conspecifics.6, 7, 8 The benefits of such color vision capabilities may have played an important role in the evolution of non-human primates into humans. Thus, although color is not a physical prop0.6 erty of the world, and considerable color perception 0.5 variation exists among humans, the ability to perceive color in the environ0.4 ment seems evolutionarily important.
autumn 2009
magnetic spectrum. The contrast encoding of receptor excitation is essential because a photoreceptor whose sensitivity distribution peaks around 540 nm only communicates the varying presence stimulating light to the brain, not its wavelength. It only says “I’m responding, I’m responding!” it does not communicate “I’m responding and I’m greenish!” The “greenish” part of the message comes when signals from different photoreceptor types are subsequently compared, beyond the retina.
volume 2.3
classes of reflectance spectra, and their variation vis-à-vis observer’s photopigments, give strong evidence that profiles of light reflected off objects are not uniquely colored. Indeed, object reflectance spectra are only electrical and magnetic pulses of photon energy waves, which do not contain any color, or even have any visual features. Thus, color is an internal construction.
The genetic basis of color vision.
A
s mentioned, color stems from object reflectance spectra, through comparisons of different photoreceptor class signals that arise from the probabilistic capture of reflected photons from a usable portion of the electro-
0.2
0.0 400
500
600
700
Nanometers
Figure 2. Two curves showing metameric reflectance spectra under a standard observer model. Both reflectances produce the same lavender color appearance (Munsell color chip 5P 6/8) shown approximated by the inset circle. The lavender screen sample of Munsell 5P 6/8 was rendered Aug. 26, 2009 at www.myperfectcolor. com/Match-of-Munsell-5p-6-8-p/mpc0110461.htm. Special thanks to Professor A. Kimball Romney for providing the reflectance spectra for use here
Figure 3 (above). Several known and estimated variations in human retinal phenotypes linked to variation in photopigment opsin genes. Curves illustrate the responsivity of different photoreceptor classes to the electromagnetic spectrum.[46] Top row depicts known observer models, bottom row depicts estimated observer models. Panel (a) shows a normal trichromat observer with short- (SWS), medium- (MWS) and long(LWS) wavelength sensitive photopigment classes; (b) a deficient dichromat, a form classically known as “Daltonism” (a Deuteranope-type missing MWS photopigment)[19]; (c) an anomalous trichromat (Deuteranomalous with shifted MWS photopigment); (d) a retinal tetrachromat with two LWS pigment classes in addition to the usual SWS and MWS photopigment classes; (e) a retinal tetrachromat with two MWS in addition to SWS and LWS photopigment classes; and (f) a retinal pentachromat with two MWS and two LWS photopigment classes in addition to the SWS photopigment. Uncertainty and debate exist regarding the phenotype expression of forms (e) and (f).
light levels; (b) the chromosome-7 linked short-wave sensitive cone photopigment; and (c) the X-chromosome linked middle- and long-wave sensitive cone photopigments.10 Genes for the X-chromosome linked photopigments are the basis for our color sensitivity at the mid- and long-wave portions of the visible spectrum, M-opsins and L-opsins, respectively, and share 98% gene sequence similarity.11, 12, 13 The structure and function of X-linked opsin genes reveal much about their evolutionary purpose as a highly adaptive component of the visual system. Several genetic features support this idea. First, considering naturally occurring genetic variations, the ability to differentiate appearances of predominantly long-wavelength frequencies from medium-wavelength frequencies arose in our primate ancestors via straightforward X-linked gene duplication – a key process in evolving new gene functions. Second, a single missing or different amino acid (called “SNP” for single-nucleotide polymorphism) in certain portions of the opsin gene sequence produces dramatic shifts in the visual response to light.14, 15 And third, duplication, divergence, intra- and inter-genic cross-overs and unequal recombination are all normal operating procedures for M- and L-cone opsin genes.
These opsin gene features contribute to differences in retinal photopigment response properties. Figure 3 shows typical retinal photopigment responses (a), compared to several variations (b-f). The initial identification of opsin gene sequences yielded unexpected M- and L-opsin gene variation.16 Subsequent research found many M- and L-opsin gene sequence variants are systematically linked to the peak responses of photopigment absorption curves. Measuring spectral response properties of different photopigment variants in vivo is complicated by varying optical density of pigments, cell “wave-guiding” morphology, and ocular media filtering. Nevertheless, variations in color vision phenotypes are traceable to genetic variation, so it’s viable to use individual opsin genotypes to investigate behaviors associated with phenotype variation. Interestingly, the X-linked inheritance of these photopigments implies that some women have different long-wavelength sensitive opsin genes on each X-chromosome and, consequently, the genetic potential to express more than the usual three photopigment classes (see online supplement at www.
glimpsejournal.com/2.3-KAJ.html#1). These heterozygous females are among those considered putative retinal tetrachromats17 and may express (in addition to rods) four retinal cone classes, each with a different spectral sensitivity distribution, thus having the potential to experience tetrachromatic vision.18 Individual variation and color perception experience.
M
volume 2.3
uch human color perception research has explored the impact of individual differences in photopigments on color perception. Response curves of observer types in the top row of Figure 3 are well-understood. Figure 3 shows (a) a normal Trichromat; (b) a deficient Dichromat; and (c) an Anomalous Trichromat. Types (b) and (c) are measurable color perception deficiencies.
autumn 2009 Color
Figure 3 (d), (e) and (f) show less well-understood forms of normal individual variation that approximate phenotypes which in theory could arise due to expressed opsin gene variation. Demonstrating such types in vivo is difficult due to considerable response similarity among the photoreceptor classes. However, their existence has been described in several studies.17, 18, 20, 21, 22 Existence of type (d) individuals with four distinct retinal cone classes is now generally acknowledged, even if types (e) and (f) are still debated.13 Type (d) is key here, and is referred to as a retinal tetrachromat.22, 23
85
Figure 3’s message is that flexibility in the structure of the X-linked opsin genes facilitates change in the genetic basis for human color vision. This same flexibility is widespread across species12 perhaps suggesting that evolving opsin gene variety itself poses no inherent evolutionary disadvantages. What do individual differences imply for emergent tetrachromacy?
T
he observer modeled in Figure 3(d) is a retinal tetrachromat, and possibly a functional tetrachromat23 who might experience color perception differences compared to a normal trichromat, and could exhibit non-normative color processing behavior on certain color perception tasks18 (see online supplement at www.glimpsejournal.com/2.3-KAJ.html#2). Figure 4 simulates some color perception differences arising from variations shown in Figure 3(a-c) and illustrates that under such observer variations object color is clearly observer-dependent and cannot belong to the object.
Understanding these normal individual differences and color vision deficiencies24 help us appreciate: (1) the extent of natural variation in color perception, (2) how little such differences have mattered historically with respect to color utility, and (3) the implications for emerging tetrachromacy at both observer- and species-level.
Figure 4. Illustration approximating the appearance of a United States Flag for color vision normals and some color deficient observers. Courtesy of the National Archives (University of Wisconsin, Americana collection, 1437652)
Neitz et al. suggested that “extra pigment types in people with normal color vision are sufficiently different to support tetra- or even pentachromacy,” but like most early researchers, downplayed the possible effects of retinal tetrachromacy, further stating, “The fact that they don’t indicates that the trichromacy of normal vision has its origin at a level of the visual pathway beyond that of the cone pigments, likely beyond the receptors.”25
www.glimpsejournal.com Glimpse
Recent results demonstrating perceptual consequences of retinal tetrachromacy complicates this accepted model of trichromacy. A three-channel post-receptor processing constraint would eviscerate new information available through retinal tetrachromacy by reducing it to a trivariant signal. Observing that signals from additional photoreceptors get used, and yield variation in perceptual experience, therefore requires an update to accepted theory. This remains a topic of considerable debate. Is the idea of potential human tetrachromacy really so strange?
S
ince the discovery of allelic variants of human long-wavelength and medium-wavelength photopigments, there’s been a good deal of popular speculation about the implications for color perception. For example, the Financial Times, “Weird Science” section explored “Women Who Can See More Than Red” (March 10, 2001, p. 11).
86 The potential for human tetrachromatic color perception need not be spun into a Sci-Fi fantasy of beings with supranormal vision. In fact, opsin gene diversity within primate species, and the natural adaptive flexibility of opsin gene structure and function, both foreshadow a real potential for human tetrachromacy in the evolutionary pipeline. Already, evidence of tetrachromacy exists in a number of animal species (see online supplement at www.glimpsejournal.com/2.3-KAJ.html#3). While most mammals are dichromats, three to five photopigments are otherwise common. At the upper extreme are mantis shrimp which seem to make use of eleven different photopigments.26 Responding, in part, to environmental changes, formerly trichromatic fish species have evolved several extra photopigments in as short as 1-2 million years, and this is linked to species’ opsin gene diversity driven by evolutionary selection pressures27 (see online supplement at www.glimpsejournal.com/2.3KAJ.html#3a). Opsin gene diversity and flexibility is also seen in non-human primates. Some New World primates exhibit considerable opsin gene diversity within species28 (see online supplement at www. glimpsejournal.com/2.3-KAJ.html#3b). Old World primate studies comparing human and chimpanzee opsin genes suggest an ongoing processes of gene conversion for some human photopigment opsin genes29, 30
(see online supplement at www.glimpsejournal.com/2.3KAJ.html#3c). And advances using transgenic therapy have transformed dichromat primate individuals into trichromats, permitting otherwise unexperiencable color sensations, and demonstrating that rapid, dramatic changes are possible in the primate neural coding of color31, 32 (see online supplement at www.glimpsejournal.com/2.3-KAJ.html#3d). But isn’t human trichromacy already optimized for our environment?
S
hepard describes human trichromacy as the most effective way to visually process and encode terrestrial light.33 However, considering that many other terrestrial animals need more than three functional photopigment classes, the optimality of the human system feels anthropocentric. Additional complications come from species with more than three photopigments operating in spectral ranges not hugely different from humans. The European Starling’s color discrimination performance, for example, suggests that at least some of the bird’s three photopigments couple with a fourth (that peaks in the near UV) within the range of 400-700 nm (see online supplement at www.glimpsejournal.com/2.3-KAJ.html#3e). Thus, in the human visible range, Starling tetrachromacy is a viable form of color processing.34 What are the selection pressures that might cause tetrachromacy to emerge?
T
he possibility of human tetrachromacy raises two intriguing considerations: (1) what visual processing demands provide positive selection pressure for tetrachromacy? and (2) what would color vision be like for a tetrachromat? We don’t know the answers to either question, but recent investigations of putative female tetrachromats are places to start. Research has found color perception differences (albeit, subtle) in comparisons of possible tetrachromat women with trichromat controls. Rigorous psychophysical studies of potential tetrachromat color perception exist,20, 21, 35 but are equivocal on the precise variation experienced under retinal tetrachromacy. Limitations in display technology and stimulus presentation formats may have historically hindered demonstration of such differences, if they exist, using traditional psychophysical methods.23 Investigations attempting to avoid such obstacles employ increased stimulus complexity and more
T
he foregoing gives clues concerning how human tetrachromacy might prove advantageous today, but we can’t predict which kinds of presentday color judgments herald behavioral advantage for the long-term. It’s possible that early non-human primate mutations in the gene structure may have been largely due to selection pressure from the environment, whereas more recent mutations may be additionally driven by social and sexual forms of evolutionary pressure. Under changing circumstances, several future evolutionary scenarios are plausible: Interpretation of human emotion states.
Moreover, Sayim et al.36 found that in some portions of color space individuals with tetrachromat genotypes shared, as a group, cognitive color-similarity representations and a color linguistic code with higher consensus (compared to trichromat controls), perhaps reflecting color expertise among such individuals. Such findings may suggest why individuals vary greatly regarding color judgments in art, publishing, architecture and design.
Changizi et al.41 suggested trichromacy evolved to detect important physiological states using color correlates of blood-oxygenation levels among nonhuman primates. So too, color correlates of emotion states might be important cues for successful social interaction and appropriate interpretation of emotion expressions among human conspecifics.42
Finally, concerning color categorization research, one might think the existence of specialized groups of color observers in a population would create problems for a population’s evolution of a shared color naming and categorization system. That is, if subsets of a society’s individuals use different perceptual categories for identifying objects, how can successful communication occur among all members, and how could a shared color communication system evolve?
Disease detection.
We used computer simulation approaches from evolutionary game theory to investigate such questions using simulated color category learning scenarios.37, 38, 39, 40 Our results showed no obstacles to evolving stable categorization solutions in populations that include agents modeled with normal, deficient and putative tetrachromat discrimination data. Indeed, some aspects of population observer
Historically, color perception has been important in medicine. Medical practitioners note red in a rash, yellowness of jaundice and the colors of a healthy body.43 Modern day doctors use color stains in cell histology and color codes on medical instruments.24 Color deficient doctors may miss symptoms because of an inability to perceive the color of disease. Informally we observed results that, although unpublished, are consistent with the idea that tetrachromacy may inform us about the uses of color in evolving technologies, for example, in medical diag-
Color
Such results imply real world consequences for individuals with extra opsin genes. For example, Jameson et al.17 suggests that one color vision test widely used in industry and the military can inadvertently classify putative tetrachromats as deficient when they may actually have richer color perception experience.
Speculations on a future for tetrachromacy...
autumn 2009
These results show that when color judgments were made in empirical conditions that approximated more naturalistic viewing circumstances (e.g. binocular viewing of contextualized largefield stimuli), processing variation correlated with human tetrachromacy was easier to demonstrate. But more specifically, the results show that the genetic potential to express more than three cone classes correlated with differences in color categorization, naming and color similarity judgments.
diversity actually help color categorization systems form and stabilize in simulation scenarios.37, 38 If analogous to color category evolution in real world linguistic societies, these results suggest that no significant communication obstacles would be expected from societies comprised of realistic proportions of normal, dichromat and tetrachromat individuals, each with varying forms of color perception and potentially different salient color categories for object identification and communication.
volume 2.3
naturalistic color processing conditions and behaviors. 17, 22, 36 These investigations used molecular genetic methods to identify potential retinal tetrachromats and found differences in perceptual behaviors when the genetic potential for more than three photopigment classes was present. Behaviors differentiating tetrachromat genotypes from trichromat controls included: (1) perceiving more colors in diffracted spectra;22 (2) correlation between performance variation on a standardized test for trichromacy and indices of richer color experience;17 and (3) color similarity and color naming pattern variation found in shared group consensus measures from potential tetrachromats compared to female trichromat controls.36
87
www.glimpsejournal.com Glimpse  
nosis. We found that laboratory assistants with tetrachromat genotypes gave different pathology diagnoses when certain cell stain scales were used in histology studies. Figure 5 (b) shows an example of the type of fluorescent cell marker (with adjacent color scale), that when used to color code panel (a) produced different estimates of cancer cell detection by potential tetrachromats compared to observers without such potential.
Figure 5. Digitized image (b) at right translates the fluorescent intensity of each cell in image (a) at left using the chromatographic scale shown at right. The scale shows that the brighter the cell, the more protein expressed. Raters count the number of bright, medium and faintly stained cells. The potential problem is that detection of the medium cells is likely to be different for raters possessing opsin gene polymorphisms (i.e., putative tetrachromats) compared to those who do not.
88
Figure 6. Multidimensional information coding: panel (a) shows the one dimensional gradient code used in monochrome (green) brightness code. Panel (b) shows two dimensions of a combined brightness and color code. Panel (c)--showing a sonar signal beam in monochrome code--and Panel (d)--showing a signal in a brightness-color code-show the two forms of the displays tested in visual processing circumstances experienced by sonar scope operators in U.S. military applications and college undergraduates.44
Co-evolutionary social pressures along these lines may have served in the past, and could serve in the future, as factors encouraging human tetrachromacy. Processing color in contextually rich information displays. Using color to identify objects involves combining different types of information, or perceptual dimensions, during information processing. While trichromacy gives greater color discrimination, studies show that color deficient dichromats may be better at detecting targets in color camouflage.44 Dichromats do this by picking out targets using luminance differences that get drowned out for trichromats by the chromatic content they appreciate. Such signal processing is a consideration for modern information displays, because while display designers want to simultaneously present all sorts of information, not all observers can easily interpret multidimensional display codes. Jameson and colleagues examined whether a one-dimensional brightness code typically used in sonar applications (Figure 6(a)) could be combined with a second dimension of color code (Figure 6(b)) to add an extra layer of information to the standard data display.45 They found that normal trichromat observers could extract two forms of information from the 2-dimensional display codes on par with the 1-dimensional code performance. Thus, observers (i) reliably detected slightly brighter signal beams in the multicolored panels (Figure 6(d)), while (ii) correctly identifying information conveyed by color in the same display (e.g., whether a signal was primarily reddish, greenish, or yellowish). Dichromats would find task (i) easy, whereas task (ii) would be difficult for a dichromat with red-green confusion. This ability to extract two forms of information from a combined code exemplifies how color dimensions could be easily separated under human tetrachromacy. It’s unclear whether in contextually-rich scenes tetrachromacy might permit identification of signal dimensions overlooked by trichromats when displayed information encodes two, three or
four dimensions of data. Obviously any such tetrachromat performance difference may be both subtle and might apply for some portions of the color space but not others.
he story of photopigment evolution suggests
On the one hand, one can entertain the possibility that human tetrachromacy reflects an on-going, natural, evolutionary potential for human visual processing. If a need arose in our environment (like dramatic environmental changes seen during the Cretaceous period, or a highly valued social trend that established a uniform color bias) human photopigment genes would be ready to meet the challenge.
Human Potential for Tetrachromacy by Kimberly A. Jameson is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License. View the online supplement to this article with additional illustrations and information at www.glimpsejournal.com/2.3-KAJ.html This PDF document is optimized for Adobe Acrobat Reader 9.0 (available for free download at http://get.adobe.com/reader) for best display and hyperlinking. Hyperlinks in this document can also be selected, copied and pasted into a browser address bar.
References 1.
Mollon, J.D., “Tho she kneel’d in that place where they grew,” The Journal of Experimental Biology 146, 21-38 (1989).
2.
Osorio, D. and Vorobyev, M., “Colour vision as an adaptation to frugivory in primates,” Proceedings of the Royal Society B 263, 593–599 (1996).
3.
Regan, B.C., Julliot, C., Simmen, B., Vienot, F., Charles-Dominique, P. and Mollon, J.D., “Fruits, foliage and the evolution of primate
On the other hand, a narrower approach focuses on sensationalizing deviations in perceptual experience brought about by tetrachromacy compared to trichromacy (cf. the Financial Times “super-perceiver” perspective noted earlier). Of these, the view of human tetrachromacy as a natural evolutionary potential seems more useful. Considerable color perception variations already exist among and within dichromat, anomalous trichromat and normal trichromat observer groups without major behavioral consequences or evolutionary meltdowns. Further research should show human tetrachromacy to
colour vision,” Philosophical Transactions of the Royal Society B 356, 229-283 (2001).
4.
Dominy, N.J. and P.W. Lucas, “Ecological importance of trichromatic vision to primates,” Nature 410, 363–366 (2001).
5.
Lucas, P.W., Dominy, N.J., Riba-Hernandez, P., Stoner, K.E., Yamashita, N., Loría-Calderón, E., Petersen-Pereira, W., RojasDurán, Y., Salas-Pena, R., Solis-Madrigal, S., Osorio, D. and Darvell, B.W., “Evolution and function of routine trichromatic vision in primates,” Evolutionary Anthropology 57, 2636-2643 (2003).
Color
human tetrachromacy may be in the pipeline. Visual pigments of vertebrates evolved about 500 million years ago (mya). Precursors of modern day human visual pigments were likely an adaptation that began in the Cretaceous period, around 150 to 80 mya. The flexibility of the opsin gene structure has permitted adaptive changes in the past, and is almost certainly ready to adapt if needed in the future. There’s no need to assume that an evolutionary zenith is realized in modern human photopigment opsin genes. If the human species survives long enough, some selective advantage, or form of co-evolution, may provide strong a justification human color vision tetrachromacy. There seem to be distinctly different ways to think about emergent human tetra-chromacy:
autumn 2009
T
volume 2.3
Summary
be correlated with subtle individual color perception variations that are no more problematic than those that already exist. While the impact of potential human tetrachromacy may turn out to be important for some applications, in general, its impact is likely to be nonproblematic at the practical levels of modern life. Even so, tetrachromacy is highly significant for theories of perception and theories of human evolution. w
89
6.
Sumner, P. and Mollon, J.D., “Colors of Primate Pelage and
17. Jameson, K.A., Bimler, D. and Wasserman, L.M., “Re-assessing Perceptual
Skin: Objective Assessment of Conspicuousness,” American
Diagnostics for Observers with Diverse Retinal Photopigment Genotypes,”
Journal of Primatology 59, 67-91 (2003).
In Progress in Colour Studies 2: Cognition, Pitchford, N.J. and Biggam, C.P., eds., (Amsterdam: John Benjamins Publishing Co., 2006)13-33.
7.
Changizi, M.A., Zhang, Q. and Shimojo, S., “Bare skin, blood and the evolution of primate colour vision,” Biology Letters, 2(2),
18. Mollon, J.D., “Worlds of Difference,” Nature 356, 378-379 (1992). 19. Dalton, J., “Extraordinary facts relating to the vision of colours: with obser-
8.
Fernandez A.A. and Morris M.R., “Sexual selection and trichro-
vations,” Memoirs of the Literary and Philosophical Society of Manchester 5,
matic color vision in primates: statistical support for the
28-45 (1798).
preexisting-bias hypothesis,” American Naturalist 170, 10-20 (2007).
20. Nagy, A.L., MacLeod, D.I.A., Heyneman, N.E. and Eiser, A., “Four Cone Pigments in Women Heterozygous for Color Deficiency,” Journal of the
9.
Sparkes, R.S., Mohandas, T., Newman, S.L., Heinzmann, C.,
Optical Society of America 71, 719-722 (1981).
Kaufman, D., Zollman, S., Leveille, P.J., Tobin, A.J. and
Glimpse
www.glimpsejournal.com
217-221 (2006).
McGinnis, J.F., “Assignment of the Rhodopsin Gene to Human Chromosome 3,” Investigative Ophthalmology & Visual Science
21. Jordan, G. and Mollon, J.D., “A Study of Women Heterozygous for Color Deficiencies,” Vision Research 33, 1495-1508 (1993).
27, 1170-1172 (1986).
22. Jameson, K.A., Highnote, S.M. and Wasserman, L.M., “Richer Color 90
10.
Nathans, J., Thomas, D. and Hogness, D.S., “Molecular genetics
Experience in Observers with Multiple Photopigment Opsin Genes,”
of human color vision: The genes encoding blue, green, and
Psychonomic Bulletin and Review 8(2), 244-261 (2001).
red pigments,” Science 232, 193-202 (1986).
23. Jameson, K.A., “Tetrachromatic Color Vision,” Invited contribution to The 11.
Jacobs, G.H. and Nathans, J., “The Evolution of Primate Color
Oxford Companion to Consciousness, Wilken, P., Bayne, T. and Cleeremans,
Vision,” Scientific American, 32-39 (April 2009).
A., eds. (New York: Oxford University Press, 2009), 155-158.
12. Jacobs, G.H., “Primate color vision: A comparative perspective,” Visual Neuroscience 25, 619-633 (2008).
13. Gegenfurtner, K.R. and L.T. Sharpe, “Color Vision: From Genes to Perception,” (Cambridge University Press, 1999).
24. Birch, J., Diagnosis of Defective Colour Vision, (London: ButterworthHeinemann, 2001).
25. Neitz, M., Neitz, J., and Jacobs, G.H., “More than three different cone pigments among people with normal color vision,” Vision Research 33, 117-122 (1993).
14.
Winderickx, J., Lindsey, D.T., Sanocki, E., Teller, D.Y., Motulsky, A.G. and Deeb, S.S., “Polymorphism in red photopigment underlies variation in colour matching,” Nature 356, 431-433
26. Kelber, A., Vorobyev, M. and Osorio, D., “Animal colour vision – behavioural tests and physiological concepts.” Biological Reviews 78, 81-118 (2003).
(1992).
27. Carleton, K., “Cichlid fish visual systems: mechanisms of spectral tuning,” 15.
Sanocki, E., Lindsey, D.T., Winderickx, J., Teller, D.Y., Deeb,
Integrative Zoology 4, 75-86 (2009).
S.S. and Motulsky, A.G., “Serine/alanine amino acid polymorphism in the L and M cone pigments: Effects on Rayleigh
28. Jacobs, G.H. and Deegan, J.F., II, “Polymorphic New World monkeys with
matches among deuteranopes, protanopes and color normal
more than three M/L cone types,” Journal of the Optical Society of America
observers,” Vision Research 33, 2139-2152 (1993).
A, 22, 2072-2080 (2005).
16. Nathans, J., Piantanida, T.P., Eddy, R.L., Shows, T.B. and
29. Verrelli, B.C., Lewis, C.M. Jr., Stone, A.C. and Perry, G. H., “Different
Hogness, D.S., “Molecular genetics of inherited variation in
Selective Pressures Shape the Molecular Evolution of Color Vision in
human color vision,” Science 232, 203-210 (1986).
Chimpanzee and Human Populations,” Molecular Biology and Evolution 25, 2735-2743 (2008).
30. Verrelli, B.C., Tishkoff, S.A., “Signatures of selection and gene conversion
41. Changizi, M.A., Zhang, Q. and Shimojo, S., “Bare skin, blood and
associated with human color vision variation,” The American Journal of
the evolution of primate colour vision,” Biology Letters 2(2),
Human Genetics 75, 363-375 (2004).
217-221 (2006).
31. Mancuso, K., Neitz, J., Hauswirth, W.W., Connor, T.B. and Neitz, M., “Gene
42. Yasuda, M., Webster, S., and Webster, M., “Color and facial
therapy treatment of color blindness in adult primates,” Journal of Vision 7,
expressions,” Journal of Vision 7(9), 946, 946a. [Abstract |
15a [Abstract 15, | doi:10.1167/7] (2007).
doi:10.1167/7.9.946] (2007).
43. Spalding, J.A.B., “Medical students and congenital colour vision
Mauck, M.C., Neitz, J. and Neitz, M, “Gene therapy for red–green colour
deficiency: Unnoticed problems and the case for screening,”
blindness in adult primates,” Nature [Advance online publication |
Occupational Medicine 49, 247-252 (1999).
44. Morgan, M.J., Adam, A. and Mollon, J.D., “Dichromats detect 33.
Shepard, R.N., “The perceptual organization of colors: An adaptation to
colour-camouflaged objects that are not detected by trichro-
regularities of the terrestrial world?” In The Adapted Mind; Evolutionary
mats,” Proceedings of the Royal Society: Biological Sciences 248,
Psychology and the Generation of Culture, Barkow, J.H., Cosmides, L., and
291-295 (1992).
45. Jameson, K.A., Kaiwi, J. and Bamber, D.E., “Color-Coding 34.
Smith, E.L., Greenwood, V.J. and Bennett, A.T.D., “Ultraviolet colour percep-
Information with Psychological-Constant Hue Loci: Assessing
tion in European starlings and Japanese quail,” The Journal of Experimental
Alternative Coding Schemes using Independent Brightness and
Biology 205, 3299-3306 (2002).
Hue Dimensions,” Journal of Experimental Psychology: Applied 7(2), 112-128 (2001).
35.
Hood, S.M., Mollon, J.D., Purves, L. and Jordan, G., “Color discrimination in carriers of color deficiency,” Vision Research 46, 2894-2900 (2006).
46. DeMarco P., Pokorny J. and Smith, V.C., “Full-spectrum cone sensitivity functions for X-chromosome-linked anomalous
36.
Sayim, B., Jameson, K.A., Alvarado, N. and Szeszel, M.K., “Semantic and
trichromats,” Journal of the Optical Society of America A: Optics
Perceptual Representations of Color: Evidence of a Shared Color-Naming
Image Science and Vision 9, 1465-1476 (1992).
Function,” The Journal of Cognition & Culture 5, 427-486 (2005).
37. Jameson, K.A. and Komarova, N.L., “Evolutionary models of color categorization. I. Population categorization systems based on normal and dichromat Observers,” Journal of the Optical Society of America A, 26(6), 1414-1423 (2009).
38. Jameson, K.A. and Komarova, N.L., “Evolutionary models of color categorization. II. Realistic observer models and population heterogeneity,” Journal of the Optical Society of America A, 26(6), 1424-1436 (2009).
39. Komarova, N.L. and Jameson, K.A., “Population Heterogeneity and Color Stimulus Heterogeneity in Agent-based Color Categorization,” Journal of Theoretical Biology 253, 680-700 (2008).
40. Komarova, N.L., Jameson, K.A. and Narens, L., “Evolutionary Models of Color Categorization based on Discrimination,” Journal of Mathematical Psychology 51, 359-382 (2007).
Color
Tooby, J., eds. (New York: Oxford University Press, 1992) 495-531.
autumn 2009
doi:10.1038/nature08401] (2009).
volume 2.3
32. Mancuso, K., Hauswirth, W.W., Li, Q., Connor, T. B., Kuchenbecker, J.A.,
91
Glimpse  
www.glimpsejournal.com
Singed Bedroom, Weekend Afternoon
92
I painted the walls plum and hung sheer Curtains so when they caught fire from the atoms Rushing from my body this afternoon It was the loveliest thing I had seen The rain came down like a song as I was Disintegrating seamlessly all electric soft colors I turned into something solar and crackling Watching from my twin bed How I wanted to reach out to my own going As a spirit might want to examine itself in a photo Barely present in a spot of faded yellow light Looking hard squinting and asking Is that me
-Arto Vaun
Photo Courtesy of Flickr Member Stacy Michelle
volume 2.3 autumn 2009 Color 93
Watercolor Science:
Transparent watercolor through the eyes of an aerospace engineer by Christie Marie Bielmeier All Images Courtesy of Dr, William L. Ko
S
tep into a NASA engineer’s office and you will find a wipe-board with mathematical equations, a library of thick reference books, a powerful computer and a collection
of engineering plans. However, step into Dr. William L. Ko’s office and you will be overwhelmed by watercolor images of train cars, railroad ties and women’s flowing skirts. Ko, a nationally acclaimed Taiwanese watercolorist and NASA scientist, is a juxtaposition of scientific excellence and artistic imagination. Perhaps it is this struggle that makes his soulful landscapes grip viewers with their precision and beauty.
www.glimpsejournal.com Glimpse 94
“To be a successful painter in transparent watercolor depends in large measure on the artist’s ability to dominate the medium. In other words, he must be able to employ a wide spectrum of techniques in order to express his emotional ideas with freedom and proficiency—as in eloquent speech,” wrote Ko in a 1976 American Artist article. “Each brush stroke must be certain, final and should transmit the very soul of the painter.” Control, restraint and practice are not what viewers first notice about Ko’s paintings, which are vibrant and mottled images. However, working under Ko for two years, this diligence is very evident in his engineering work. Since 1977, working as an aerospace engineer for NASA, Ko has published over 100 technical articles on a broad range of topics including space shuttle tile heating, aircraft structure loading, and innovative designs for hypersonic aircraft. It is this experience —and a lightning-fast ability to crunch numbers with pencil and paper—that commands respect. His colleagues at NASA Dryden in Mojave, California honor Ko by calling him simply “Dr. Ko” in an otherwise casual office.
(previous page, below) Illustrations from Dr. William L. Ko, US Patent #4,292,375 for Superplastically formed diffusion bonded metallic structure. (left, following page) Watercolor paintings, by Dr. William L. Ko, from Railroad Short Stories and Railroad Fine Art (2008). (below, right) Watercolor study by Dr. William L. Ko. All images courtesy of Dr. William L. Ko.
volume 2.3 autumn 2009 Color 95
I first met Ko in 2003 after graduating from engineering school. He is a quiet, compact man of more than eighty years and is always neatly dressed in a collared shirt and bolo tie. He is a man of few words, but is known to hand out freshly picked peaches and whimsical drawings of Korean characters. This warm man was my first engineering mentor on my first real-world engineering assignment.
Uncertain how to proceed, I entered the complex model into the computer and received wrong results—again and again! Frustrated, I sought out Dr. Ko. His approach to solving the mathematical problems is best detailed by his explanation of the merits of using transparent watercolor.
Straight out of college, I thought engineering involved shoving variables into high-tech computers and then waiting for the box to spit out complicated mathematical solutions. However, completing a stress analysis on space components is a detailed and complicated undertaking. Ko is an expert in this field and does everything by hand—for good reason!
“If all watercolors were perfectly transparent, the first brush stroke on a white paper will produce but one color. Next, if we paint the second color on the paper—partly overlapping the first color after it has dried—we can generate at most three colors.” Ko continues this process by using three brush strokes to produce seven colors and five brush strokes to produce fifteen colors. This process creates a series of numbers known as geometrical progression. To me, this process illustrates the buildup and layering of information to produce a complex result— whether artistic or mathematic.
I was assigned to determine what force would break an aircraft’s wing spar, or circular hollow tube. This was no ordinary tube: it had an oval cross-section made of several different materials. My first instinct was to use the computer, but software programs require models of standard circular tubes made of a single material. I started to calculate the problem by hand, but there were too many variables.
To tackle my stress problem, Ko advised me to start small and then add complexity. I began by creating a model of a circular aluminum tube and quickly solved the simple math problem by hand. Then, I put the model into the computer and compared results—they matched! I used this success to continue adding complexity and different material properties. When the computer failed to match the results on my paper, I tweaked a parameter and tried it again. Each success was measured in small, exact steps, which increased confidence in my answer.
96 Glimpse   www.glimpsejournal.com
volume 2.3
autumn 2009 Color
97
Glimpse
www.glimpsejournal.com
Finally, my model became so complex that hand calculation became too difficult. I had an idea of what the approximate answer should be, but not a definitive answer. The answer was just out of focus— similar to a mottled Ko landscape.
98
Ko writes, “I enjoy working in transparent watercolor because the fresh dispersion of the color, which often dimly defines objects, gives the observer a mystical feeling.” Dimly defining the answer was Ko’s next suggestion. Using the best model that could be calculated with pen and paper, I approximated the breaking force for the best and worst cases. These results told me my computer answer had to be within a precise range. With trepidation, I tried the computer, which gave me an answer within the range. I was confident in this answer because of the foundation I had built. Ko taught me, by using diligence and persistence, the best answer can become quite clear—even when hazed by false information. “I usually practice the same subject matter repeatedly, until my brush movement is free and florid, capable of a full expression made with a limited number of brush strokes,” says Ko, who studied at the Southern School of Chinese Art. This strict schooling mandates that ideal paintings require no washing out or other correction. For this reason, he practices painting objects—countless times—prior to final execution. It is these beautiful practices that hung on Ko’s office walls. While I worked for Ko, the images tended to consist of green and brown railway connection brackets with a few shots of pink flowing fabric. These works now are part of a new book, Railroad Short Stories and Railroad Fine Art, released by Ko in 2008. Ko’s illustrations are paired with short stories written by his father, Seth Mackay Ko, and showcase Taiwanese trains. The book is a result of Ko’s short stint as an express train driver in 1950s and a lifetime fascination with railways. “My familiarity with the mechanical details of the steam locomotive greatly help my train drawings,” says Ko, who hopes the book will serve as a memorial to his father. While working with Ko, I witnessed the influence his watercolor training had on his engineering, but his engineering also influences his watercolor. Like a good research scientist, he studied his work
intensely and employed a few tools—paper, distilled water and a timer—to aid him in creating the most vivid color possible. A painting is only as good as the paper it’s painted on. Or at least that’s what Ko decided to prove with a study of watercolor paper, which tends to fade under different types of lighting. On strips of different weight papers, Ko painted watercolors of different colors. For the next six months, the strips were exposed to direct sunlight, fluorescent light, incandescent light or darkness. Ko documented that “all the color exposed under the sunlight faded negligibly, with the exception of the violet and alizarin families, which faded badly. Color exposed under the two kinds of light show hardly any fading (usually dyed paper will fade faster under the fluorescent light than under incandesce light). After this test, I have avoided use of unstable colors on my palette.” One painting that will not fade in importance is LBJ Ranch in Spring Time, which Ko painted for President Johnson in 1968. It has been in the permanent collection at the Lyndon Baines Johnson Library in Austin, TX since 1975. During the mid-1960s, Ko showcased his talents as a one-hour watercolor master at world’s fairs across the
1927 Born in Tamsui, Taiwan 1950 Mechanical Engineering BS from Taiwan University Worked as machine design engineer for Taiwanese Railroad Administration 1963 Aerospace Engineering PhD from California Institute of Technology 1977 NASA Engineer
2008 Publishes Railroad Short Stories and Railroad Fine Art,
1932 Began painting and received training from artists now at the National Taiwan Academy of Arts. 1959 Aerospace Engineering MS from California Institute of Technology 1968 Paints “LBJ Ranch in Spring Time” for President Lyndon Baines Johnson 1995 NASA Exceptional Service Medal
country. This one-man show was invited to the LBJ Ranch where Ko painted for the president. Ko says, “I visited the LBJ Ranch several times and saw the ranch covered with yellow wild flowers. This was my inspiration.”
Color
Watercolor painting allows Dr. Ko to build up from the white surface, a precise solution to the visual problem. Perhaps that is why he is drawn to the medium. Overall, Ko feels his lives as a painter and engineer are interconnected. “Painting is a lifelong process of gathering beauties of God’s creations. Engineering is a lifelong thinking process of inventing new technologies.” w
autumn 2009
To make sure his paper stays wet, Ko likes to work at a humidity of 70 percent, which is hard to come by in the desert where he lives. To add moisture to
Creating the highest amount of color is a passion for Ko, which is why he enjoys watercolors. He feels that other opaque color mediums, such as oil, limit the number of potential colors—allowing only one color per brush stroke. To him, non-transparent media are an inefficient use of his brush stroke, since watercolor is three times more effective at creating color. However, he does admit that watercolors tend to become cloudy when too many colors are added, which makes the number of independent colors only theoretical.
volume 2.3
His one-hour painting is not simply a carnival trick. It is based on research of paper wetness and its affect on color boldness. “I usually start painting on the wet surface about ten minutes after the surface has been wetted. Usually, the wet paper will last for less than an hour depending on the atmospheric humidity. If the painting cannot be finished within the first period of wetting, rewetting (after the paper has completely dried) is conducted for applying the finishing touches. However, this is not recommended, since rewetting will always dissolve the colors already painted to some degree.”
the air, Ko lets his lawn sprinklers run. He says that wetting his lawn and then painting near the grass allows his paper to stay damp for up to two hours.
(below, left; below, right) Watercolor studies by Dr. William L. Ko. All images courtesy of Dr. William L. Ko.
99
(Re)Views By Ivy Moylan
BLUE
Glimpse
www.glimpsejournal.com
Derek Jarman, (1993) 79 min.
100
In many ways Blue is an anti-film. Not in that it is against film but, like an antihero, it is the opposite of a film while still being one. You watch it, you listen to it, and it engages you as an audience member. But it is simply a blue screen with narration. No cinematography, editing, image composition. None of the normal elements that one thinks of, and discusses about, a film. But director Derek Jarman did intend Blue to be a film. Therefore, it is a film. Experimental, but a film nonetheless. I started hearing about Jarman’s Blue when Kieslowski’s Blue came out—they were released around the same time and it was somewhat confusing for film audiences and exhibitors; and since I really enjoyed Jarman’s Edward II and am grateful for his discovery of the amazing Tilda Swinton, I was curious, but admittedly intimidated and a bit afraid of it. Would I even get it? How hard is it to watch a screen that never changes, is only blue, for nearly an hour and a half? Not as hard as I expected. The narration is beautifully written. Jarman was partially blind when he made the film, only four months before he died from AIDS-related complications, and the single wash of color mixed with narration that emulates thought and experience gives the viewer an opportunity to experience the world in a different way, but a way that feels natural. This was one of the first things that I considered while watching the film, that this could be a representation of what the world was like to Jarman at the end of his life: which is full of experiences, ideas and thoughts, but missing the images. The narrator shares experiences—possibly memories—as well as his internal musings. We sit with him sharing where he is right now—a café with some friends—and we hear the noises of the café combined with the narration. Then he begins sharing with us one of his doctor’s visits. The narration continues to travel, skipping, like any of our thoughts do, from one idea to the next. From beach, to hospital, to a memory. And we just listen, imagine, and think. Watching Blue, I thought. I thought a lot. Between the narration and the lack of imagery, there was actually a lot to think about. I spent a lot of time thinking about what a movie consists of. Because Blue contains only the elements of sound, sound editing, and one image/color, the absence of imagery in fact provided me the opportunity to meditate on the idea of film and the multitude of information that I absorb while viewing. The writing is very visual and the sound design also helps flesh out these images, so I felt that I could see much of what he was describing. And in seeing it only in my imagination, I thought about what I would see if it were actually onscreen—if it was actually what the movie was—once again, making me super-aware of the film devices that could possibly be used. Then, of course, I also thought about color. The word “blue” is peppered throughout the narration. Every time it was used, I was brought back and noticed what I was actually watching—a blue screen—but was able to stay connected to the sound because I was seeing and hearing the word “blue.” Other colors are mentioned in the film as well: magenta, red, yellow. Although I was looking at a blue screen, I was able to imagine these colors with vividness equal to the blue onscreen. Although I was just hearing the word, I felt very aware of color. It is a surprising relationship between the evocative narration and the static screen. One I could never have imagined without watching the film. Blue is poetic and very intimate. It is heartbreaking and vital. Yet, it is just a blue screen. w
THE WIZARD OF OZ
Victor Fleming (1939) 101 min. Starring: Judy Garland, Frank Morgan, Ray Bolger, Bert Lahr, Jack Haley, Billie Burk, Margaret Hamilton.
My own memory of the film seems to be affected by its use of color as well. Every time I watch The Wizard of Oz, I am surprised at how much happens in the film before Dorothy ever gets to Oz. Not only do we meet all of the main actors of the film, but she sings “Over the Rainbow,” the most famous song from the musical! Because the jaw-dropping moment when Dorothy opens her door into Oz is such a shocking color transition, the earlier part of the movie fades into the background of our memories, the same way its color palette pales in comparison to the Technicolor Munchkinland that Dorothy now sees. Oz is just a cacophony of color: Ruby slippers, Emerald City, Yellow Brick Road. The possibility that the world can be in color, instead of a washed-out sepia, makes the viewer believe that anything else could be possible there, too. So the fact that the film becomes a musical or that there are talking trees and dancing tin men, or that there are good and bad witches, is no more unbelievable than that the world is colorful and bright, or as Dorothy decides, “on the other side of the rainbow.” Because The Wizard of Oz was shown annually on television starting in the 1950s, it has become a central piece of American film history, and has left an indelible mark on nearly all American childhoods. Dorothy is a wonderful coming-of-age heroine, with bravery, intelligence, and faith to lead her trio of misfits to the Emerald City and to their hearts’ desires. I also like the lesson, visually-emphasized in the film through color, that we all have the power within ourselves to reach whatever goal we desire, whether it be a brain, heart, courage or just getting home. w
(left) Blue, courtesy of flickr member Sleeping Sun. (this page) Yellow Brick Wall, courtesy or flickr member Matti Mattila. (right) Red Shoes, courtesy of flickr member Lily Bozier.
Color
One of the reasons that The Wizard of Oz might be mistaken as the first color movie, is because color is such an intrinsic part of the film. From the sepia-toned Kansas (shot in black and white and then colored sepia), to the green of Emerald City, color is a central character in this film. It is what carries the story along, literally, the Yellow Brick Road. Many of us were raised on black and white TV and it was a revelation when we first saw The Wizard of Oz in color. As Roger Ebert wrote on www.rogerebert.com, “It was not until I saw The Wizard of Oz for the first time that I consciously noticed B&W versus color, as Dorothy was blown out of Kansas and into Oz.”
autumn 2009
It is commonly thought that The Wizard of Oz was the first color feature film, which it wasn’t. But it was the first color feature film made by MGM studios. And it is one of the most well-known early Technicolor films, along with it predecessors The Adventures of Robin Hood and Walt Disney’s Snow White and the Seven Dwarves, as well as Gone with the Wind, which was also released in 1939 and also directed by Victor Fleming.
volume 2.3
Technicolor. Although the word is now part of American vernacular, its origin is from a film company by the same name, founded in Boston, Massachusetts, that invented the Technicolor process that brought us brighterthan-life color films from the 1920s through the 70s (Godfather 2 was the last Technicolor film, made in 1974). It is a labor-intensive process where all three primary colors are shot onto separate film reels and then combined to make a film with intensified colors.
101
In the next issue...
Cosmos
*
Glimpse  
www.glimpsejournal.com
Katharina Lodders The Chemical Elements of the Cosmos
*
Jason W. Barnes Seeing Titan: Mapping Saturn’s moon with infrared technology
102
*
Kimberly A. Jameson and Jon Lomberg The Use of Color in Interstellar Message Design
* *
Michael R. Molnar What the Wise Men Saw in the Sky
Scott Kardel Dimming the Lights: Astronomy and light pollution
*
and much more..........
*
Ross A. Beyer Exploring Mars and The Moon in Google Earth
Questions of Anthopology
Key Textbook!
Edited by Rita Astuti, Jonathan Parry and Charles Stafford
Anthropology today seems to shy away from the big, comparative questions that ordinary people in many
societies find compelling. Questions of Anthropology brings these issues back to the centre of anthropological concerns. Individual essays explore birth, death and sexuality, puzzles about the relationship between science and religion, questions about the nature of ritual, work, political leadership and genocide, and our personal fears and desires, from the quest to control the future and to find one’s ‘true’ identity to the fear of being alone. Written in an engaging and accessible style, Questions of Anthropology presents an exciting introduction to the purpose and value of Anthropology today.
LSE Economics Monographs on Social Anthropology July 07 • 288 pp PB 9781845207489 £16.99 HB 9781845207496 £55.00 LSE Monographs - ISSN 0077-1074
With over 70 volumes published since 1949, including classic works by Gell, Barth, Leach and Firth, the LSE Monographs now form one of the most prestigious series in the discipline of Anthropology. Presenting scholarly work from all branches of Social Anthropology, the series continues to build on its history with both theoretical and ethnographic studies of the contemporary world. order online at w w w.bergpublishers.com
The
Skin Quilt Project
In the upcoming documentary, The Skin Quilt Project, we discover the hidden secrets of colorism, and its lasting effects within the African American community. Join with the producers of The Skin Quilt Project as they interview African American students, quilters, scholars, and artists on the relationship between the African American quilting tradition, the perception of skin color, and appreciating one’s cultural heritage.
For more information contact: info@skinquiltproject.com www.skinquiltproject.com
The MIT Press Synthetic Times MEDIA ART CHINA edited by Fan Di’an and Zhang Ga Innovative and groundbreaking works by new media artists from nearly thirty countries reflect what it means to be human on the threshold of human-machine symbiosis. Copublished with the National Art Museum of China 358 pp., 200 color illus., $44.95 paper
Working-Class Network Society COMMUNICATION TECHNOLOGY AND THE INFORMATION HAVELESS IN URBAN CHINA Jack Linchuan Qiu foreword by Manuel Castells afterword by Carolyn Cartier “Qiu has written the most insightful, empirically-grounded account to-date of the social role that the Internet and related information and communication technologies have played in the course of China’s rapid economic development. Anyone with an interest in the social and economic implications of the Internet in developing economies.” — William H. Dutton, University of Oxford Information Revolution and Global Politics series 320 pp., 25 illus., $35 cloth
To order call 800-405-1619 t http://mitpress.mit.edu t Visit our e-books store: http://mitpress-ebooks.mit.edu
Explore China’s Architectural Heritage Open Hearts Open Doors Reflections on China’s Past and Future ELIZABETH GILL LUI
AFTERWORD
BY
I. M. PEI
“Lui’s photographs of rural China transform photojournalism into visual poetry. The places they feature are living spaces imbued with feeling and memory. Seemingly timeless, these beautiful images also convey a subtle sense of transformation, telling a familiar legend with a fresh voice.” —WU HUNG, UNIVERSITY OF CHICAGO 250 PAGES / $50.00 CLOTH | 160 COLOR PHOTOGRAPHS
Available in two bilingual editions— English/Simplified and English/Mandarin
FOUR STOPS PRESS DISTRIBUTED BY CORNELL UNIVERSITY PRESS
1-800-666-2211 www.cornellpress.cornell.edu
volume 2.3
autumn 2009 Color
105
106 Glimpse   www.glimpsejournal.com