Painting Perception

Harvard scientists try to dissect the enigma of art

Whitney E. Adair

What more can science uncover by turning its gaze on art­—or, conversely, what can art teach the scientists?

Writing not long after the death of Leonardo da Vinci, art historian and biographer Giorgio Vasari described the late master’s “Mona Lisa,” placing special emphasis on the lady’s uncanny simper. “And in this work of Leonardo’s there was a smile so pleasing, that it was a thing more divine than human to behold; and it was held to be something marvelous, since the reality was not more alive,” he wrote. The sublime expression of “La Joconde” has held sway over its viewers since its creation, a testament to its creator’s artistry, a manifestation of the fantastic in the earthly realm.

Enter Dr. Margaret S. Livingstone, Professor of Neurobiology at Harvard Medical School, whose research focuses on human visual perception. Livingstone realized that while contemporary art historians like Ernst Gombrich are not wrong in their analysis of “Mona Lisa,” there’s a science to da Vinci’s masterpiece that had yet to be fully explained. Analyzing the work in terms of its spatial frequencies, Livingstone revealed that the lower spatial frequencies, best seen by the peripheral vision, make the figure appear to smile, while at higher frequencies the smile almost vanishes. Laying one famous enigma to rest, however, calls up a host of other questions: what more can science uncover by turning its gaze on art­—or, conversely, what can art teach the scientists? And just how important is a foundation in one field for an understanding of the other?

BLURRING THE LINES

Livingstone is far from alone in her exploration of the science animating works of art. Many of her colleagues, including some at Harvard, pursue similar interests; they channel their curiosity about human visual perception into an artistic study or use scientific findings to explain some of the fundamental principles that underlie works of art. The Vision Sciences Laboratory, located in William James Hall and run by a group of psychology professors, explores this very chiasm in their experiments—though their approach is slightly different than Livingstone’s.

“You might think that the Vision Lab works on the software more than the hardware,” says Patrick Cavanagh, Professor of Psychology and Co-Director of the lab. Diverging from Livingstone’s approach, which Cavanagh relates to disassembling an Xbox to figure out what’s going on inside, the Vision Sciences Laboratory conducts tests on individual subjects in order to ascertain when and how their powers of perception fall short. By presenting subjects with pictures, actions, and events, the researchers hone in—be it through brain scanning or more simple tests of visual judgments—on perceptual errors, and induce the workings of the brain.

Cavanagh’s own research, which takes the same approach, has led him to a number of discoveries about an “alternative physics” operating in representational artworks. Impossible shadows, reflections, and contours number among the artistic flaws surveyed in an article Cavanagh wrote for the science journal “Nature” entitled “Artists as Neuroscientists.” “Artists use this alternative physics because these particular deviations from true physics do not matter to the viewer,” he writes. “The artists can take shortcuts, presenting cues more economically, and arranging surfaces and lights to suit the message of the piece rather than the requirements of the physical world.”

Through a careful survey of many works of art, confirmed by experiments on perception, Cavanagh has discovered that the only requirements for realistic depictions of shadows are that they be transparent and darker than their immediate surroundings. He has shown that reflections, like shadows, are a mystery to the human mind; their representation in art has but a few of the limitations which govern reality. He maintains, in a similar vein as Livingstone, that Impressionist art is so appealing because intentional blurring may connect representations more directly to emotional centers in the brain rather than to conscious image-recognition areas. Cavanagh has even offered an explanation why flat, two-dimensional representations are effective, arguing that we do not experience the visual world as truly three-dimensional.

Implicit in all these findings is the understanding that the artists themselves must have had some knowledge of these very rules in order to replicate them in their works. And their comprehension of the weak points in human visual perception has become a boon to modern researchers. “You can figure out from artists and what they do just what the simple rules of vision are,” Cavanagh says. “And that’s a real advantage, it’s like lots of research done for free.”

Livingstone had a similar revelation early in her career. “I began to realize that artists, just like scientists, are doing experiments, and that a number of works of art reveal important things about how we see,” she says.

SYMBIOSIS OR PARASITISM?

Even while artists act like scientists and scientists like art historians, it is not easy to decipher the role the sciences should play in the expansive field of art history. While Cavanagh maintains that an understanding of human visual perception is helpful and informative in—if not essential to—the study of art, Livingstone understands that barriers do exist. While some scientific developments, like fractal analysis, have already swept the world of art history, she claims, others may not be accepted so easily by art historians. “Anything innovative takes time to be understood by the establishment, especially if it takes them outside their areas of expertise and comfort,” she writes in an email.

Indeed, some are less convinced of the merits of the scientific approach to artistic interpretation. “There’s a methodological problem at the core of the entire approach, and it has to do with consciousness and subconsciousness,” says Frank Fehrenbach, Professor of History of Art and Architecture. “What are you testing? What is your primary material? All these people [researchers] start with vocal response from their test persons. If you develop ideas about how art works on us then you determine that art must work on us in a certain way.”

Fehrenbach, familiar with the advances in neurobiology and psychology and what they offer art historians from a seminar he recently taught, says that a more science-oriented approach is a prominent channel in art history that seems to be gaining momentum. His concerns with it, however, range from the methodological issue mentioned above to more fundamental areas of disagreement.

On one hand, Fehrenbach argues that asking subjects to respond to a work of art to analyze what’s happening in the brain can’t offer anything more than simple questions to an audience or to oneself. On the other hand, he stresses that this type of research-oriented approach is only a new substitute for formalism and places an undue emphasis on immediate response. Neurological or psychological studies could easily fall short in explaining such instantaneous reactions. If scientists observe that subjects have similar responses to a work of art, can they necessarily prove that similar mental mechanisms are at work beneath them?

Touching again on the methodological, Fehrenbach recalls Margaret Livingstone’s explanation of Mona Lisa’s smile. “Now someone is giving us a scientific explanation for it, and there are a few ways that you can respond to it. First of all, you can say great, finally that has been explained by scientific fact. Or you can say, well does that mean, since I do not agree with the floating emotion in the Mona Lisa, does that mean that my neurologic apparatus is not okay? … I would simply say that all that neuroscience can provide is post-festum observations of what people already perceived on a fundamental level.”

Ultimately, the biggest problem for Fehrenbach is not the research itself or its impact on art history, but the implicit promise that we will be able to develop a new art theory. “It’s impossible to create a set of categories that would allow you to understand what art is or how it works on us,” he says.

Despite even these gravest of doubts, Fehrenbach admits that in certain areas the approaches taken by Cavanagh, Livingstone, and the Vision Sciences Laboratory in the interpretation of artworks could be a fertile foundation for greater understanding. These areas, however, are not particularly numerous; nor do they initially seem to offer the same boundless opportunity that researchers like Cavanagh envision in the confluence between art history and the science of perception.

COMING TO A COMPROMISE

Neither side in this discussion advocates a takeover or excision of one field from the other. To a certain extent, both seek to understand the membrane between art and science—the liminal area between art and science that has spawned many a discussion, particularly as the sciences continue to grow beyond the confines of their laboratories and into the cultural mainstream. Cavanagh, in discussing whether a background understanding of visual perception can inform the study of art, acknowledges that it certainly depends upon what the individual in question is interested in studying. He gives precedence to an understanding through symbolism and cultural context in many instances. Fehrenbach, too, occupies a bit of this middle ground: “To put the sensual part of art back on the agenda is absolutely important,” he says. “Art theory sometimes tends to contextualize too much, but it should start with the perceptual qualities of the object.”

—Staff writer Joshua J. Kearney can be reached at jkearney@fas.harvard.edu.

Tags