The Competitiveness of Nations in a Global Knowledge-Based Economy


The biology of art

The Economist, April 3, 1999, 69-71


Art for science’s sake

The cortex of the beholder

Come up and see my etchings


Few areas of knowledge escape scientific scrutiny these days—even those often thought of as being at the opposite end of the intellectual spectrum

SCIENCE is generally the farthest thing from people’s minds when they take in the beauty of a Van Gogh or a Monet.  And by the same token, artists rarely consult scientists before painting a canvas.  But it was not always so.  During the Italian Renaissance, when many of the conventions of western art were first established, what is now called science frequently informed the way that people painted.  And some artists went farther, not only using the new knowledge of geometry and anatomy to improve their art, but also speculating about how and why art was perceived in particular ways.

Leonardo da Vinci, that archetypal Renaissance man, conceived of what he called the ten functions of the eye.  These were darkness, light, body, colour, shape, location, distance, closeness, motion and rest.  He also came up with the idea that although the images of objects travel physically from the front of the eye to the imprensivci (now called the retina), they are actually formed in the sensus communis (the imagination or the brain).

Those were bold guesses.  But as neurophysiologists have investigated the way in which vision works, they have found that Leonardo was right on both counts.

On top of that, over the past few years, they have discovered ways in which the tricks employed by artists exploit these divisions of labour to create illusions of reality.  That provides some answers to the question “how?”  Neither is the question “why?” being ignored.  Evolutionary biologists are now asking themselves just what art is for.  And some are concluding that it is not merely an accidental manifestation of human intelligence, but a specific, evolved capacity with a clearly defined purpose.


Art for science’s sake

When a beam of light is focused by the cornea and the lens, at the front of the eye, on to the retina, at the back, it stimulates photosensitive cells known as rods and cones.  Rods are sensitive to all wavelengths.  Cones, which are responsible for colour vision, come in three varieties, each tuned to a particular wavelength.

A stimulated rod or cone emits an electrical impulse, which in turn stimulates an impulse in a nerve cell.  This travels into the brain along an optic nerve (there is one for each eye) to arrive at one of two small clusters of nerve cells called the lateral geniculate bodies.  These are split into two layers, the magnocellular and parvocellular systems.  The role of the magno cells is to signal contrasts in brightness, while the parvo cells react to both brightness and colour.  After that, the divided signal is channelled for analysis to the visual cortex, a large, convoluted region at the very back of the brain.

It is here, in the visual cortex, that artists play their games.  For it is here that the signal is really chopped into Leonardo-like categories before it finally emerges, by processes as yet poorly understood, as what the possessor of the brain in question perceives as visual reality.

The visual cortex is divided into some two dozen areas, but the most thoroughly investigated are those that handle the incoming signal first.  These are named, rather prosaically, VI, V2, V3, V4 and V5.  Each has a specific function or functions.  The first, V1, has three subdivisions of cells, each of which processes a particular part of the signal from the magno or parvo cells, and then sends the result to the second area, V2.

One of the leading exponents of research on these areas is David Hubel, of Harvard University.  A few years ago his colleague Margaret Livingstone realised that the results might be brought to bear on the question of artistic technique.

To determine which visual signals are the most important to each subdivision of V1, Dr Hubel’s group (among others) has measured the electrical responses of its nerve cells to a range of stimuli.  The differences between individual cells are striking.  Some are activated by colour or brightness but not at all by shape or movement.  Others are sensitive to orientation but not to colour or movement.

The researchers then turned to V2 (the next region to receive the signal) and exposed it to the same stimuli.  By following the path of the signal electrically - and viewing slices of V2 tissue under the microscope - they discovered that V2 also has three subdivisions, each of which receives its input from one of V1’s subdivisions.

The role of the other three regions, V3, V4 and V5, is the province of another group that is interested in the science of art.  This group works at University College London under the tutelage of Semir Zeki and his colleague Matthew Lamb.

Initially, Dr Zeki employed a scanning technique called positron-emission tomography to investigate these regions.  PET uses radioactive oxygen to measure blood flow (and therefore activity) in different parts of the brain.  PET scanning showed Dr Zeki that regions V3 and V5 are particularly sensitive to movement, an observation corroborated by the fact that people who sustain damage to V5 (from a stroke, for example) suffer


from so-called motion blindness.

As with other areas of the visual cortex, the sensitivity of V3 and V5 cells is specialised.  V3 cells respond to diagonally oriented lines, especially when they are in motion. V5 cells react to motion in general, with each cell responding particularly strongly to motion in one preferred direction.  Both V3 and V5, however, are indifferent to colour.  That “function of the eye” is the province of V4.  So if a stroke or tumour damages a person’s V4 region, he will become completely colour-blind.

The brain’s interpretation of colour is actually a three-step process.  First, the wavelength-composition of the light from all points in the visual field is analysed and registered by V1 and V2.  Second, this information is sent to V4, where the light reflected from a surface is compared with that reflected from surrounding surfaces, in order to control for relative rather than absolute brightness.  Third, the results of this analysis are relayed back to V1 and V2 for reinterpretation, after which the signal is sent to the parts of the brain where the impression of an image seems to emerge.

Put all these observations together, and visual processing seems to follow three separate, but intertwined, pathways through particular regions of V1 and V2 and then on to V3, V4 or V5.  The first path runs from the magno cells via V1 and V2 to V5.   The second runs from both magno and parvo cells via V1 and V2 to V4.  And the third runs exclusively from the parvo cells, via V1 and V2 to V3.  According to Dr Zeki and Dr Lamb, this analysis goes a long way towards explaining how different art movements achieve their particular effects.


The cortex of the beholder

The kinetic art movement that began around 1914 - when artists felt the urge to incorporate the illusion, and ultimately the reality, of motion into their work - was the first artistic school to be analysed by Dr Zeki and Dr Lamb.  Artists’ first attempts at experimenting with motion were crude, as in Duchamp’s “Nude Descending the Staircase”.  This collage of coloured human forms only feebly suggests motion, since the emphasised colours and forms stimulate V4, while the motion-sensitive V3 and V5 regions are ignored.  True artistic exploitation of motion neurobiology did not appear until the 1930s, with Calder’s mobiles.  These, of course, moved.  But by abandoning colour entirely, and making his mobiles black, Calder unknowingly minimised the activation of the colour pathway while optimally stimulating V3 and V5.

In Harvard, a different artistic school attracted the researchers’ attentions.  Dr Livingstone looked at the late-19th-century impressionists, such as Seurat, who were playing their tricks on other parts of the visual cortex.

According to Dr Livingstone, although the individual brush-strokes or dots of impressionist paintings are large enough to be seen by the predominantly form-sensitive third pathway, they are too fine to be detected by the colour-sensitive second pathway.  So, when such a painting is observed from a distance, the colours blend.  Up close, however, the dots are large enough to stimulate those few form-sensitive nerve cells that are part of the second pathway, and are thus perceived as separate.  The result is that the colours no longer blend.  (This effect may be noted in the living room as well as the art gallery.  It is what allows you to watch television without seeing the individual “pixels” that make up the picture.)

Op-art represents yet another artistic flirtation with neurophysiology.  For example, Mondrian’s painting “Broadway Boogie Woogie” is widely perceived as unstable, as if it is jumping all the over the canvas.  According to Dr Livingstone, the colour combinations here (yellow lines against an off-white background) are strong stimuli for pathway three, the shape detector, and weak activators of pathway one, which looks for edges but cannot reliably find any between these two colours and so cannot fix the position of what it is looking at.

Dr Zeki’s most recent foray is into fauvism, an art movement whose aim was to shock the viewer with vibrant colours.  He decided to explore the colour pathway beyond V4, to work out which brain areas are used when objects are seen in their natural colours and when they are seen in the wrong colours.  A vivid example of this is André Derain’s “Charing Cross Bridge”, which re­


although the river does not look like water (it is yellow - not a natural colour, even for the Thames), it nonetheless is just that.

For these experiments, Dr Zeki used a second brain-scanning technique, functional magnetic-reso­nance imaging (fMRI).  This jiggles hydrogen atoms into giving off radio waves that can be turned into a picture of a brain (or, indeed, any other part of a body).  Since most of the molecules of which the brain is made contain hydrogen - but do so in varying quantities - fine-grained maps of the organ can be built up by looking at the hydrogen density, and changes in blood flow tracked rapidly.  It is therefore possible to see which areas are active during specific tasks.

Dr Zeki and his colleagues asked people to view scenes of fruits, vegetables, animals and landscapes in both their natural colours and in abnormal colours.  The fMRI results showed that naturally coloured objects stimulate four areas of the brain that lie beyond V4.  These regions are known to have sophisticated roles, such as memory-storage, in the brain’s integration of the various “functions of the eye” into a meaningful visual experience.  However, none of them was activated when the people inside the fMRI machine were shown pictures of objects in the wrong colour.

There is also comfort for those who wonder about the aesthetic value of much of modern art.  In a separate fMRI experiment Dr Zeki discovered that multi-coloured abstract paintings, like Mondrian ‘s, stimulate the brain up to area V4 but not beyond it.  Representational art, by contrast, goes all the way.  By reducing the features of the visual world to simpler elements, modern artists have also minimised the number of visual pathways used to view their work. That simplification surely cannot mean that it is more easily appreciated by simple minds.  Can it?

Come up and see my etchings

“IS IT art?” is a question that is asked not only by critics.  Archaeologists have an interest in the matter, too.  Mankind’s ancestors have been fabricating artefacts for at least 2.5m years.  But for most of that time, those artefacts were merely undecorated stone tools.

The flowering of ancient art happened quite recently.  The oldest known artistic manifestations in Western Europe - one of the world’s better explored crannies - date from about 35,000 years ago.  This was the time when modern people were replacing Neanderthals in that part of the world.  Art objects in the form of jewellery (beads and pendants) began to appear and, most spectacularly, the cave paintings of Spain and France began to be composed.  Art had arrived with a vengeance.

That pattern is repeated elsewhere.  It is the appearance of modern man, Homo sapiens, that coincides with the appearance of art.  The older species of man that were displaced by him showed little inclination to pick up a paint brush.

Why that was so is a matter of debate.  Some scholars believe art is merely an accidental manifestation of a change in human cognitive processes.  Modern people, this school of thought contends, “see” the world differently from their predecessors.  This cognitive difference permits them to make, and embroider, external representations of that world - in other words, to create art.

It may well be true that such a cognitive change occurred.  Certainly something must explain the success of Homo sapiens, and the speed with which the species spread and swept other hominids away.  Since there are few anatomical differences between human species, cognitive ones are the most likely cause of modern man’s predominance.  But to permit something is not to ensure that it happens.  And a second line of reasoning leads to the idea that art has quite a specific function that of showing off to the opposite sex.

The originator of this theory is Geoffrey Miller, of University College London.  He believes that the capacity to produce art evolved in a way similar to that famous piece of natural frippery, the peacock’s tail.

This process is known as sexual selection (to distinguish it from the more general process of natural selection).  Natural selection eliminates organisms from the gene pool by killing them.  Sexual selection acts even more directly.  It eliminates organisms from the gene pool by preventing them from finding a mate.  Sexual selection emphasises the development of characteristics that demonstrate in a way that cannot be faked just how fit that characteristic’s possessor is to father (or mother) another individual’s offspring.

Sometimes the characteristics are subtle.  Though people are generally unaware of it, one of the most important things that makes them (and many other species) attractive to the opposite sex is bodily symmetry.  Perfectly symmetrical growth is difficult to achieve.  So someone who is exactly symmetrical probably has very good genes indeed.

Often, however, the characteristic is spectacularly gaudy - like the peacock’s tail.  No unhealthy bird could possibly grow such an appendage, so it is a clear sign of good genes.

According to Dr Miller, art is like this.  A number of important skills are involved in producing it.  Decorative art in almost every human culture includes precisely repeated patterns such as stripes, checks, stars, lattices and knots.  Any deviation from perfect repetition in such a pattern is noticed easily.

Before the age of mechanical reproduction, artists would have required great skill, dedication, time and energy to produce such exact repetitions.  Just as it is only genetically fit peacocks that can grow symmetrical tails with perfectly repeated eye-spots, only genetically fit artists, with good muscular co-ordination and perfect vision, can produce accurate repetitions of complex, symmetrical motifs.

Representational art also tests these skills - and, in addition to them, taxes the imagination and intelligence of the artist.  Humans, like many mammals, are known to be attracted to novelty for novelty’s sake.  That is probably why originality is so highly valued in art (and therefore why reproductions - even high-quality ones  - are less esteemed than the work from which they have been copied).  To produce an unimaginative or derivative work of art, therefore, will not have as desirable an effect on potential partners.

Co-ordination, vision, imagination and intelligence.  An individual with all of these is likely to make a good mate - in genetic terms, at least.  Whether an artist of genius can abide the pram in the hall is a different question entirely.