The eye is the window to the soul.

—Leonardo da Vinci

mong all his scientific and artistic studies, Leonardo allotted disproportionate attention to light and the body’s detector of light, the eye. He experimented with the optical phenomena of reflection and refraction, created reams of drawings of geometric optics in order to understand the behavior of light incident on multifaceted as well as uniformly curved surfaces, and performed exquisite anatomical dissections on the eye itself. Unhappily, he accorded little attention to the structure and function of the brain (unless such work was presented in his missing manuscripts, for we only have one-third of the original volume he produced). When he created his astonishing portraits, the eyes of the subjects spoke (or withheld) volumes in accordance with its artist’s own vision and perfect artistic control. The ambiguity portrayed in the Mona Lisa’s mesmerizing visage represents Leonardo’s insight into light and optics, as well as psychology.

The quality of art one produces is as much a function of how one observes as of how one wields a pencil, a paintbrush, or a chisel. Observing is a collaboration of the eye sensing and the brain processing. In childhood, drafting skills continue to improve until we are around ten years old.1 But then 95 percent of the population begins to display regression in their drawing prowess. Rather than focusing on their subject—that is, consciously observing—they summon subconscious preconceptions and transfer to their drawing boards what they think their subject looks like. By their middle years, in the unlikely case that they actually pick up a pen and pad to sketch the face of an individual, they have generally fallen into the habit of placing their subject’s eyes higher and higher in the face. The eyes belong on the equator. The remaining 5 percent continue to improve, drawing what they observe. They comprise the roughly one person in twenty whom we characterize as possessing some measure of artistic ability. In teaching nonartistic adults how to draw, Betty Edwards in a modern classic, Drawing on the Right Side of the Brain,2 suggested a remedy: take a photograph of the subject and mount it upside down. Then, draw what you see.

Workshops using Edwards’s ideas endeavor to teach students techniques enabling them to access the nonverbal, spatially oriented right brain, best suited for observing and drawing. The question that arises here is whether there is any evidence of physiological and functional differences in the brains of artists and nonartists.

The Painter’s Eye3

That artistic talent could be detected scientifically might have sounded absurd until a series of experiments in the 1990s revealed that perhaps it is not such a laughable notion after all. The gifted English representational artist Humphrey Ocean and a personal friend, scientist John Tchalenko, collaborated in a series of studies. As reported by writer Alan Riding, while Tchalenko and his camerawoman, Belinda Parsons, were videotaping Ocean, Ocean was creating a portrait of them.4 In the early 1990s the documentary film produced by Tchalenko and Parsons, along with Ocean’s artwork, were shown in exhibitions at a number of leading galleries in the United Kingdom. Near the end of the decade, Tchalenko revisited the time-encrypted videotapes produced almost a decade earlier and discovered a pattern in Ocean’s artistic technique previously overlooked. The discovery inspired Tchalenko to undertake further experimentation.

In 1998 he drafted neuroscientists from Oxford University and Stanford University who used state-of-the-art biomedical imaging technology. These new experiments on Ocean focused on the interactions between his eyes, hand, and brain. Moreover, they involved his conscious as well as unconscious responses. These results were compared with test results of nonartists performing similar tasks and were presented together in 1999 at the National Portrait Gallery exhibition The Painter’s Eye. The three scientific sections of the exhibition carried the titles “The Eye Captures,” “The Hand Implements,” and “The Brain Processes.” “The Eye Captures” experiment examined the actual process of observation exercised by the artist and was performed by the team headed by Christopher Miall of the physiology department at Oxford. The regularities and other patterns Tchalenko had observed in viewing his old tapes were in fact what had fueled this aspect of the new study. Using a special eye-tracking camera, the researchers performed measurements on the artist’s eye movement while he was selecting a subject and while he was actually executing his art. In choosing a subject from four candidates, Ocean was found to perform rapid eye fixations, as often as 140 times in one minute, focusing on each candidate’s features—the nose, the eyes, the countenance, and other abstract requirements that the artist might envision for his work. In commencing with the drawing he altered his pace, glancing up at his subject on the average of 12 times per minute, taking mental snapshots of one-second duration. Thus 20 percent of his working time was spent in observing his subject. During the five hours Ocean spent producing his drawing, one hour was invested in looking at his subject. When nonartists were tested, no regular patterns in their observations were found. Their fixations were simply far less methodical in manner.

The second experiment, “The Hand Implements,” examined eye-hand correlation as Ocean toiled on a sketch. He had been given a finite amount of time—just twelve minutes—to complete his drawing of a subject’s face. A sensor measured motions of his pencil on and off the paper and determined that his hand would practice strokes with his pencil just millimeters above the paper before actually landing the pencil on the paper to create a line. As Ocean characterized this procedure, “In all my work, I’m after precision. ‘Detail’ means where the line lands, and if it lands a millimeter to the right or a millimeter to the left, it changes the weight, in some way, or the shape that it is describing. So when the line lands, you want it to land in the right position, whatever that is.”5

The third section of the exhibition, and the one that involved the latest biomedical imaging technology, also revealed the most unexpected differences between the brains of the artist and the nonartist. The Oxford researchers had made quantifiable correlation in hand-eye activity for the artist and nonartist, but producing brain images was not part of their experiment. Ocean and Tchalenko were to find the last bit of their puzzle at Stanford University, where functional magnetic resonance imaging (fMRI) tests were carried out on Ocean and two nonartists by Robert Solso and his team. Each of the subjects, including Ocean, was provided with small sketch pads and a set of five abstract figures to reproduce on their sketch pads while they were lying in the confines of the machine.

With fMRI one can pinpoint the areas of the brain where various mental functions are being carried out. Blood rushes to the parts of the brain requiring oxygen to process a particular function, a particular operation, performing a mathematical calculation, or creating a work of art. The fMRI studies on artist Humphrey Ocean and on the nonartist are wonderfully revealing. During the drawing sessions in which the nonartist made drawings while inside the machine, the parts of his brain involved in accomplishing the task were revealed to be located primarily in the rear of the brain, in the visual cortex. The brain process for the nonartist consisted of seeing and copying, but distinctly did not involve abstract thought. Producing art involves the process of abstraction. When artist Ocean was working in the confines of the fMRI the areas of his brain that “lit up,” or where oxygen-carrying blood saw a surge, were found to be in the frontal part of the brain. This presents a crucial difference. In Tchalenko’s words, “For Humphrey, the real transformation was taking place in the front, where you find emotion, previous faces, painting experience, intentions and so on. In essence, the control subject was simply trying to slavishly copy what he saw. But Humphrey was creating an abstracted representation of each photograph. He was thinking the portraits.”6 And according to neurophysiologist Robert Solso, for Ocean there was “less activity in the face area, the fusiform gyrus in the right parietal area, than in the nonartist, but more activity in the right frontal areas—the ‘thinking’ part.”7

Humphrey Ocean is an immensely talented figurative artist with no fewer than five paintings (and two drawings) hanging in London’s National Portrait Gallery.8 Tchalenko, although neither a physiologist nor a psychologist (he is a trained seismologist), is certainly familiar with scientific methodology. As for those teams of scientists at Oxford and Stanford, they operate two of the finest laboratories of their kind in the world. The limitations in the scope of the experiment notwithstanding—studies on a single artist—the results are fundamentally simple and intuitive, though they clearly invite expanded study. Further study, however, will have to wait. Tchalenko notes that it is difficult to raise money for such an interdisciplinary art and science project, as “science bodies think this is art, and art bodies think this is science.”9

The Eye of the Beheld

If I could write the beauty of your eyes
And in fresh numbers, number all your graces,
The age to come would say, “This poet lies;”
His heavenly touches ne’er touch’d earthly faces.10

—William Shakespeare

“Is it the eyes?” wrote Guy Gugliotta, “The light? The sort of smile? What is it that makes the Mona Lisa—one of history’s most memorable portraits—so compelling, even in its creation, that the young Raphael sat at Leonardo’s knee just to watch him paint it?”11 Christopher Tyler, a neuroscientist in San Francisco, and independently, Michael Nicholls and fellow psychologists in Melbourne, compiled complementary evidence of particular symmetries and asymmetries, a lateral bias, in portraits spanning the past five hundred years.

In 1998 Christopher Tyler began to explore the possibility that the asymmetric functions of the two hemispheres of the brain may have somehow manifested themselves differently in artists’ works. It was an unusual approach, although decidedly simple in conception, in which he constructed a vertical line bisecting the frames of each of the portraits. Tyler accepted only seated or standing, but not reclining figures; in all, 282 artists were represented in his study. After he discovered that a preponderance of the center lines passed through one of the eyes of the subject, a “compositionally dominant eye,” he expanded his study to a much larger universe of Western portraiture spanning 2,000 years and ever-changing styles and schools.

For his statistical analysis Tyler tested four hypotheses: the major axis hypothesis, the golden section (divine proportion) hypothesis, the head-centered hypothesis, and the one-eye centered hypothesis. At first glance most observers would agree that the eyes in portraits are generally located near the center of the canvas, but Tyler’s study revealed a subtler element. One eye, either the leading or the trailing eye, was found to lie in a gaussian distribution (bell curve) about the center line with a narrow standard deviation of ±5 percent of the frame width. One-third of the portraits displayed one-eye coincident with the center line and fully two-thirds of the portraits displayed one-eye within 5 percent of the center line. Vertically, Tyler found, the eye height distribution peaked not around the horizontal center line, but rather around the golden ratio of 61.8 percent of the height of the canvas, with only a negligible number of eyes found below the vertical center.12

Tyler published a matrix of nine masterworks of Western civilization, portraits spanning the past five centuries (Plate 12, top). These were chosen to illustrate the variety of compositional asymmetries, yet all have an eye right on the center line. Even Picasso, the ultimate creator-rebel of twentieth-century art in his Portrait of Dora Maar (1937), succumbed to this unwitting, unwritten norm for artists to put an eye at or near the vertical center line (third row, third column). Until the end of the twentieth century, common American currency—from George Washington on the one-dollar bill to Benjamin Franklin on the hundred-dollar bill—had one eye of the subject coincident with the centerline. But then in the last years of the century new United States currency bearing values beyond one dollar were issued placing the head and the compositionally dominant eye off-center. In the self-portrait drawn by Leonardo da Vinci gazing in a mirror, an engaging exercise for the reader would be to return to the drawing, construct a vertical bisector, and view its position on the face (Plate 1, left).

Rules are made to be broken, especially if they are promulgated formally. But those unwritten rules that are generated subconsciously, perhaps inherent in the artist’s psyche, appear less susceptible to violation. And so the center line principle, falling into the latter category, has demonstrated remarkable resilience. Artists, who are otherwise compelled to excavate new ground, much like scientists, appear unlikely to stray from the principle. Nonetheless, examples of “symmetry breaking” obviously exist, a third of the eyes falling in the tails of the peak plus-or-minus one standard deviation of Tyler’s eye-distribution histogram.

Although no clear pattern outside the gaussian is discernible, ϕ, the ubiquitous ratio resonating through past chapters, is found in a contemporary painting. It appears in the work of artist Jamie Wyeth, of the distinguished family of artists. In 1967, then not quite twenty-one years old, he painted a penetrating psychological portrait of John F. Kennedy (Plate 12, bottom). On the canvas Kennedy’s face is distinctly offset from the vertical bisector of the frame,13 appearing to comply generally with the artist’s “rule of thirds.” But when a vertical line is constructed to divide the canvas 1:1.618 parts, it is seen to pass surprisingly close to Kennedy’s left eye. The canvas itself does not form a golden rectangle (possessing instead the proportion 1:1.78); thus it is the horizontal span that is divided according to the golden mean. On viewing this painting in an earlier version of my manuscript, Christopher Tyler offered an incisive comment: “It is clear that Wyeth’s intent was to create a view of Kennedy in a reflective mood, disengaged from the viewer. [The painter] has thus avoided the salience of the center line to enhance the sense of disengagement.”14 There is no evidence that the artist chose the position of the eye based on any mathematical strategy. He was simply following his artistic instincts.

A Leftward Bias

As Christopher Tyler’s serendipitous discovery indicates, the artist frequently places one of the subject’s eyes at the center line of his frame, imbuing the portrait with a hidden symmetry. But immediately the question arises: “Is there a directional bias? Is there a preponderance of one eye, one cheek—left or right—over their opposites?” Although Tyler’s eye-centered principle emerged in 1998, the issue of a directional bias had been investigated a quarter century earlier. In addition, the more recent observations by Michael Nicholls are entirely compatible with Tyler’s results and also offer compelling reasons toward resolving why the bias exists.

A 1973 paper pointed out that in a sampling of 1,474 single-subject portraits, 68 percent of female and 56 percent of male subjects presented more of the left side of the face than the right.15 Another study examining portraits by Francisco Goya alone cited a turning bias of left cheek over right cheek for female subjects, and a less pronounced disparity of right cheek over left for male subjects.16 Subsequently, an examination of 4,180 single subject portraits in a variety of media—paintings, drawings, lithographs and photographs—revealed again that an asymmetry existed, with a preference for the left cheek over the right.17 Yet another study of 127 portraits of scientists, all members of the British Royal Society, revealed no leftward or rightward bias at all.18 Moreover, a separate pair of studies found an overall preference of artists to light their subjects from the left side.19 Finally, there was Tyler’s 1998 study which led to the center line principle, pointing out that bilateral symmetry in the presentation of the face was relatively rare. With the center line passing through or near one eye, the subject was usually presented with one eye and one cheek compositionally dominant over the opposite eye and cheek, but with no lateral bias in the distribution.

With the aforementioned studies, some seemingly contradictory, serving as backdrop, Michael Nicholls carried out additional studies—some alone20 (2000) and others in collaboration with colleagues in the department of psychology at the University of Melbourne (1999). A number of issues were addressed: (1) Does the right or left-handedness of the artist influence the direction of the subject’s pose in a portrait? (2) How is the left-cheek bias reflected in a self-portrait? (3) Is there a difference in bias when the subject attempts to convey a different emotion? (4) Is there a gender factor in any of these results?

A preponderance of portraits dating from the Renaissance displays illumination of the subject from the artist’s left side (the subject’s right). Although left-handers are disproportionately represented among artists, right-handed artists still slightly outnumber their left-handed counterparts. And one would assume that artists, holding their paintbrushes in their preferred hand and their palettes in the other, would benefit from having the palette illuminated by the light falling from the illuminated side, thus reflecting the greater number of right-handed artists preferring left-hand lighting. Another hypothesis put forth by Nicholls: the right-hander, just by the musculature of his arm and shoulder, would find it more natural to draw or paint a subject with a left cheek bias, and a left-hander just the opposite. Yet an inventory of the portraits by two prolific known left-handers—Raphael and Hans Holbein the Younger—does not reflect these artists’ preference for illuminating their subjects from the left, or for displaying the right cheek/eye as the compositionally dominant one. For Raphael fully two-thirds of his single subject portraits show a left-cheek bias, for Holbein the number is three-fourths, these numbers reflecting roughly the statistics for all artists, regardless of whether they are right- or left-handed. Thus the handedness of the artist determining the side of the face featured appears to be a red herring.

Nicholls’s investigation of 137 Renaissance portraits, ignoring gender differences in the subjects, revealed that 129 of them display the center line passing through or near one eye,21 in line with Tyler’s discovery of just a year earlier. But then he found that 63 percent of the 129 portraits displayed the subject turned, exhibiting a prominent left cheek, left eye.

How does the left cheek bias exhibit itself in self-portraits? Nicholls performed statistical analyses on self-portraits as well as portraits of others, and the differences were dramatic, although in retrospect not surprising. For portraits of others, the featured side of the face was the left side in 57 percent of male subjects and 78 percent of female subjects. In contradistinction, in self-portraits it was the right side that was featured prominently—in 61 percent of portraits with male subjects and 67 percent of portraits with female subjects. These proportions for the portraits of others versus self-portraits are virtually inverted. But then, when an artist works with a mirror, the right and left sides become inverted: the left cheek becomes the right, and the right, the left! Leonardo’s self-portrait (see Plate 1) featured the right side of the artist’s face, which we again know to be his left! In the fifty-seven known Rembrandt self-portraits, those unparalleled psychological studies, spanning the Dutch master’s lifetime, forty-eight of them feature the right cheek, only nine the left. But in all those forty-eight self-portraits Rembrandt was gazing in the mirror at the left side of his face (Plate 13).

The next hypothesis Nicholls tested was the effect of the emotional frame of mind of the subject on the direction in which he or she turns. (In this experiment the premise is that it is the subject and not the artist/photographer who selects the subject’s direction of turn.) The subjects were 165 psychology students, 122 female and 43 male. An integral aspect of the study was the subjects’ initial prepping with one of two directives—actually a pair of hypothetical scenarios with differing emotional content. With neither the subject nor the proctor alerted about the left-right connotations of the study, the proctor would photograph the subject with a video camera positioned two meters in front of them. One statement prompted the subject to pose for a photograph designated for a loved one; the other prompted, “You have been invited to become a member of the Royal Society and a portrait of you is to be submitted.” After being told not to face the camera directly and to take thirty seconds to strike a pose, each subject was photographed. The first statement had been designed to generate a sense of warmth, endearment, and emotion; the second to generate a desire to suppress emotion, to exude a sense of rationality and power. Neuropsychologist Michael Nicholls offered an explanation: the right side of the brain—the hemisphere associated with emotive functions and the hemisphere controlling the left eye and cheek—impels the subject of the portrait to subconsciously turn toward the right, exposing prominently the left side. Mona Lisa, desiring to appear loving and loyal, bares her left cheek. Conversely, when immersed in the second hypothetical scenario, “to pose for a portrait to hang with portraits of other powerful and intelligent scientists in the Royal Society,” the subject of the portrait shows off the right eye (and cheek), the side controlled by the rational, expressive, confident left hemisphere of the brain. As a general principle, Nicholls’s explanation is compelling and, like the Ocean/Tchalenko experiments discussed earlier, beckons further research, if possible, with brain imaging technology. The numbers certainly suggest a correlation with differing functions of the right and left hemispheres of the brain. Moreover, the gender factor is quite minimal, and so is the handedness of the artist.