ARE YOU IMPRESSED with meals that look like one food but are actually made of something else? Tofu burgers and artificial crabmeat, for example, are not what they appear to be, yet the masquerade half-convinces our taste buds all the same.
Such ruses have a venerable history. In medieval times fish was cooked to imitate venison during Lent, when it was customary to abstain from meat and other indulgences. At all times of the year, celebratory banquets included extravagant (and sometimes disturbing) delicacies such as meatballs made to resemble oranges and shellfish made into mock viscera. Recipe books from the Middle Ages and the Renaissance also describe roasted chickens that appeared to sing, peacocks redressed in their own feathers and made to breathe fire, and a dish aptly named Trojan hog, in which a whole roasted pig was stuffed with an assortment of smaller creatures such as birds and shellfish, to the amusement and delight of cherished dinner guests. Food illusions don’t appeal only to the palate. Some exploit quirks of our neurological wiring to confuse and entertain both the eyes and mind.
Take this still life by Italian painter Giuseppe Arcimboldo (1527–1593), which depicts the ingredients for his favorite minestrone soup (top left). Turned upside down (top right), Arcimboldo’s bowl of vegetables becomes a whimsical portrait of a man’s head, complete with a serving-bowl hat.
This image raises a couple of questions. First, why do we see a face in the arrangement, when we know that it is just a bunch of vegetables? Our brains are hardwired to detect, recognize and discern facial features and expressions using only minimal data. This ability is critical to our interactions with other people and is the reason that we perceive personality and emotion in everything from crude masks to the front ends of cars.
Second, why do we see the face only when the image is flipped? The same brain mechanisms that make face processing fast and effortless are optimized to recognize faces the way we generally see them—right-side-up—so upside-down ones are harder to recognize.
A Lot to Digest
Arcimboldo’s work demonstrates that, neuroscientifically speaking, the whole is more than the sum of its parts. Clever arrangements of individual fruits, flowers and legumes become exquisite portraiture when viewed in their entirety, such as in the likeness of the Hapsburg emperor Rudolf II, here depicted as Vertumnus, the Etruscan god of transformations.
The brain builds representations of objects from line segments and tiny patches of color, then identifies them for what they are by comparing them to a mental library of similar visual images. The viewer first makes out a bulbous protrusion in the middle of Rudolf’s face because thousands of retinal photoreceptors in the eye react to the various shades of color and luminance in that area of the painting. There are no retinal cells specialized in recognizing noses, however. That next step occurs when high-level neuronal circuits in cognitive areas match the information to the brain’s stored template for noses, created from a lifetime’s experience of viewing them.
In this case, the output from those same photoreceptors also activates the high-level object-tuned neurons that recognize fruits, which is what makes images such as these so much fun to look at. A nose is a nose is a nose, to riff on Gertrude Stein—except when it’s a pear.
Such visual puns, artfully constructed, appeal to the mind as much as any wordplay. For example, in this image of a hummingbird the brain simultaneously detects animal features (eyes, wings, tail) along with plant parts (eggplant, artichoke leaves). This dual spark of recognition, with all its contradictions, tickles the fancy.
The dots that compose this image of a cherry-topped cupcake are multicolored jelly beans, a technique that recalls the works of painters such as Georges Seurat and Paul Signac. The pointillists juxtaposed multiple individual points to create hues that were very different—when viewed at a distance—from the actual colors of the painted dots.