Portrait of the interior designer Madeleine Castaing by the French painter Chaïm Soutine.
Complex causes: One great challenge in neuroscience is understanding how higher-order brain processes interact to create a viewer’s experience of art, such as this portrait of the interior designer Madeleine Castaing by the French painter Chaïm Soutine.
The Metropolitan Museum of Art / Artist Rights Society/ Art Resource

The creative brain—an edited excerpt from ‘Essays on Art and Science’

In his new book, neuroscientist Eric Kandel explores how sensory perception and higher-order cognitive processes influence our experience of art.

The greatest enterprise for the human mind has been and always will be the attempted linkages between the sciences and the humanities.

E.O. Wilson

My new collection of essays, “Essays on Art and Science,” brings together my lifelong passion for art and my life’s work in neuroscience to explore the question of how we experience art. Central to that exploration is my specific fascination with the art and culture of Vienna 1900. This was an era in which the free exchange of ideas among artists, art historians, and scientists gave rise to modernism and to the concept of the “beholder’s share”: the realization that art is incomplete without the perceptual and emotional involvement of the viewer.

From that concept, we have begun to learn how our brain uses universal, specialized rules to construct the visual world, such as the image we see in a painting. But the brain also brings to bear on that image our individual personal experiences, memories, and emotions. Understanding how these higher-order processes interact to create the beholder’s share is one of the great challenges confronting brain science in the twenty-first century. Artists themselves realize the importance of these unconscious processes—their own and the viewer’s—and may deliberately play upon them or challenge them in creating their art.


The use of strong tactile elements in a painting adds an important dimension to the beholder’s response. One of the first art historians to emphasize this was Bernard Berenson. In The Florentine Painters of the Renaissance he argued that “the essential in the art of painting . . . was . . . to stimulate our consciousness of tactile values” and thus to appeal to our tactile imagination just as much as the actual objects do. Berenson goes on to say that form—volume, bulk, and texture—is a principal element of our aesthetic enjoyment. In viewing a Giotto, for example, our visual sensations are translated into sensations of touch, pressure, and grasp. Such visual perception is a unified experience of mind and body, both physiological and psychological unity.

Perception involves a number of senses, not just vision. In evaluating the beholder’s response to art, historians have often underestimated the brain’s ability to coordinate the interaction of our senses, particularly the visual and the haptic but also, when appropriate, the sense of taste or smell. In recent years the classical idea that the brain processes the various senses separately has been replaced by a new conceptual view of a “metamodal” brain whose organization facilitates the carrying out of multisensory tasks.

Modern brain science has revealed that several regions of the cortex thought to be specialized for processing visual information are also activated by touch. One particularly important region is the lateral occipital complex, a region of the cortex that responds to both the sight and the touch of an object. The textures of objects also activate neurons in a neighboring area, the medial occipital cortex, whether those objects are perceived by the eye or the hand. This is why we can easily identify and distinguish between different materials—skin, cloth, wood, or metal—and can do so at a glance.

Brain-imaging studies have revealed that the way visual information about materials is coded changes gradually. In the early stages, visual processing of a painting or any other object is completely and solely visual. Further processing results in a multisensory representation of the object in our brain (specifically, in the fusiform gyrus and collateral sulcus) that enables us to categorize different materials. In these higher regions of the brain, perception of texture, is intimately tied to visual discrimination, because this part of the brain’s visual system has a robust and efficient set of mechanisms for processing textured images. Indeed, cross-modal association is key to the brain’s experience of art.

In addition to visual and tactile interactions, our brains can recruit powerful emotions—sometimes pleasure, but often fear, anxiety, and uncertainty—in response to paintings, such as the torturous, asymmetrical, and existential images of French painter Chaïm Soutine. In 1900 the great Viennese art historian Alois Riegl was the first to try to bridge art history and science by focusing on the psychology of perception. Riegl argued that no work of art is complete without the response of the beholder, the beholder’s share. Today we can begin to outline in a preliminary way a set of neural circuits for the beholder’s share that goes considerably further than Berenson’s important beginning. The figure below shows that in addition to visual perception of an image, including the translation of the visual experience into tactile sensations, the brain contains a representation of the face, particularly the emotion conveyed by the face, which is mediated through the amygdala, the executive structure for emotion. The beholder’s share also includes perception of the body, the body in motion, emotion, simulation, empathy, and theory of mind.

I focus here only on a few, more recently delineated components of the neural circuit of the beholder’s share. Moreover, although I schematically illustrate the various components that contribute to the beholder’s share as interacting in a linear manner, in fact they do not. They also have important additional connections, including feedback connections with one another. Higher-order cognitive areas involved in the theory of mind that have received information from earlier areas involved in visual and tactile processing can feed back to and modify the processing of earlier sensory areas involved in face recognition.

Flow diagram of the neural circuit involved in the “beholder’s share”.
Art experience: The neural circuits involved in the beholder’s share include those that process visual and tactile stimuli, emotions, movement, empathy and theory of mind.
Chris Wilcox

One interesting finding is the mirror neuron system, which is involved in imitation and was discovered in monkeys by Giacomo Rizzolatti, an Italian brain scientist. Rizzolatti realized that some cells in the motor system of the monkey brain respond not only to movements that monkeys make but also to movements that others, including people, make. This mirror neuron system, it has been suggested, may also be responsible for our empathic response to images and to works of art.

In addition to the mirror neuron system, art recruits other components of the neural circuits of the beholder’s share, including brain systems concerned with empathy and theory of mind, processes that are concerned with another person’s feelings, goals, and aspirations. Thus, the beholder not only projects his own feelings onto a person or persons in a painting but also tries to understand the goals and aspirations of the subjects. As we have seen, this also holds true for the tactile sense. The sculptural quality of Soutine’s painting actually gives us an experience of touch, and this interacts with our visual sensibility. When we “feel the texture of the paint” with our eyes, we are really re-creating the actions of the artist as he created the image. Much as mirror neurons may be used to interpret the meaning of observed behaviors, we can almost see the actions of the artist’s hand in the traces on the canvas. The notion of using thick paint to convey agitation is important; we see it also in the repetitious short, parallel brushstrokes in Van Gogh’s late work, where texture and brushstrokes represent emotions rather than people or things.

Brain areas involved in the “beholder’s share”
Interacting systems: Distinct regions of the brain contribute to the various aspects of a viewer’s response to art.
Therese Winslow

The various functions of the beholder’s share are located in distinct regions of the brain. Thus, the striate cortex, which is the first visual processing relay in the cerebral cortex, projects to the posterior and anterior temporal cortex, where faces are represented. Behind the striate cortex lies an extrastriate body area, which processes images of the body. Behind the extrastriate body area is an area that processes motion, whether generated by an automobile or a person. By contrast, in the upper part (superior) of the temporal cortex is an area that processes only biological motion, such as what occurs when a person is moving or reaching a hand out to greet you. Children with autism can respond perfectly well to the movement of a car, but have difficulty with the socially significant biological movements of a person.

In the parietal cortex and the frontal supplementary motor cortex are two areas containing mirror neurons, which are involved in imitation, as noted above. In the temporal parietal junction is an area concerned with theory of mind, part of a brain system concerned with empathy, as described above. Here, for example, the beholder tries to understand empathetically what is going on in the mind of Madeleine Castaing as depicted by Soutine. Finally, many of these areas are connected reciprocally with the amygdala, the orchestrator of our emotions.

How these higher-order processes interact to create the totality of the beholder’s share of a Soutine painting is one of the great challenges confronting brain science in the twenty-first century.

Excerpted from Essays on Art and Science by Eric Kandel, published by Columbia University Press. Copyright (c) 2024 Columbia University Press. Used by arrangement with the publisher. All rights reserved.