In other words human artists through trial and error, through intuition, through genius have discovered the figural primitives of our perceptual grammar. They are tapping into these and creating for your brain the equivalent of the long stick with the three stripes for the chick's brain. And what you end up with is a Henry Moore or a Picasso.

The advantage of these ideas is you can test them experimentally. You can actually record from cells in the brain which sort of fire when you show it a face in the fusiform gyrus. Now some of them will fire only to a particular view of a face. But higher up you've got neurons which respond to any view of a given face. And I'm predicting that if you present a Cubist portrait of a monkey face - where you present two views of a monkey's face in the same place - that cell will be hyper-activated. Just as the long stick with the three red stripes hyper-activates the beak-detecting neurons in the chick's brain, this Cubist portrait of a monkey face will hyper-activate these face-detecting neurons in the monkey brain - and the monkey says: "Wow! What a face". So what you have here is in fact a neural explanation for Picasso, for Cubism.

I've told you about one law so far - peak shift and the idea of ultra-normal stimuli. We have borrowed insights from ethology, neurophysiology, rat psychology to account for why people like non-realistic art.

The second law is more familiar to all of you. It's called Grouping.

Many of you may have seen those famous puzzle pictures, like Richard Gregory's Dalmatian dog. You just see a bunch of splotches when you first look at it but you sense you visual brain trying to solve a perceptual problem, trying to make sense of this chaos. And then after a few seconds, or maybe actually several seconds - 30 or 40 seconds - suddenly everything clicks in place and you group all the correct fragments together, and lo and behold you see a Dalmatian dog.

Richard Gregory's Dalmatian

Richard Gregory's Dalmatian

You can almost sense your brain groping for a solution to the perceptual riddle and as soon as you successfully group the correct fragments together to see the dog, what I suggest is a message gets sent from the visual centres of the brain to the limbic-emotional brain centres of the brain giving it a jolt and saying: "AHA, there is a dog" or "AHA, there is a face".

The Dalmatian dog example is very important because it reminds us that vision is an extraordinarily complex and sophisticated process. And even looking at a simple scene involves a complex hierarchy, a stage by stage processing. At each stage in the hierarchy of processing, when a partial solution is achieved - "Hey it looks a bit dog-like right here" - there is a reward signal "AHA", a partial "AHA", and a small bias is sent back to earlier stages to facilitate the further binding of the features of the dog. And through such progressive bootstrapping the final dog clicks in place to create the final big "AHA!" Vision has much more in common with problem solving - more like a twenty questions game - than we usually realize.

The grouping principle is widely used in both Indian and in Western art - and even in fashion design. For example you go to Harrods, and you pick out a scarf with red splotches on it. Then you often match it with a skirt which has got some red splotches on it. Now what's this all about? Is it just hype, is it just marketing? Or is it telling you something very deep about how the brain is organized? I'm going to argue it is telling you something very deep, something to do with the way the brain evolved.

Vision evolved mainly to discover objects and to defeat camouflage. You don't realize this when you look around you and you see clearly defined objects.

But imagine your primate ancestors scurrying up in the treetops trying to detect a lion seen behind fluttering green foliage. What you get inside the eyeball on the retina is just a bunch of yellow lion fragments obscured by all the leaves. But the brain says - so to speak - "What's the likelihood that all these different yellow fragments are exactly the same yellow simply by chance? Zero. They must all belong to one object, so let me link them together, glue them together. Oh my God, it's a lion - let me out of here!" And as soon as you glue them together, a signal gets sent to the limbic system saying: "AHA, there's something object-like, pay attention here".

So there's an arousal, and an attention which then titillates the limbic system, and you pay attention and you dodge the lion.

And such "AHAs" are created, I maintain, at every stage in the visual hierarchy as partial object-like entities are discovered that draw your interest and attention. What the artist tries to do is to generate as many of these "AHA" signals in as many visual areas as possible by more optimally exciting these areas with his paintings or sculptures than you could achieve with natural visual scenes or realistic images. Not a bad definition of art if you think about it.

That takes me to the third law - the law of perceptual problem solving or visual peekaboo. Now what do I mean by that?

As anyone knows a nude seen behind a diaphanous veil is much more alluring and tantalizing than a full-colour Playboy photo or a Chippendale pinup - or a Page Three girl, is that what you call it? Why?

As I said our brains evolved in highly camouflaged environments. Imagine you are chasing your mate through dense fog. Then you want every stage in the process - every partial glimpse of her - to be pleasing enough to prompt further visual search - so you don't give up the search prematurely in frustration. In other words, the wiring of your visual centres to your emotional centres ensures that the very act of searching for the solution is pleasing, just as struggling with a jigsaw puzzle is pleasing long before the final "AHA". Once again it's about generating as many "AHAs" as possible in your brain.

The fourth law is the law of isolation or understatement.

You all know that a simple outline doodle by Picasso or a nude by Rodin or Klimt can be much more evocative than a full colour photo of a woman. Similarly the cartoon-like outline drawings of bulls in the Lascaux Caves are much more powerful and evocative of the animal than a National Geographic photograph of a bull. Hence the famous aphorism in art: "Less is more".

But why should this be so? Isn't it the exact opposite of the first law, the idea of hyperbole, of trying to excite as many "AHAs" as possible? A pinup or a Page Three girl after all has much more information. It's going to excite many more areas in your brain, many more neurons, so why isn't it more beautiful?

The way out of this paradox is to consider another visual phenomenon, called Attention. It's a well-known fact that you can't have two overlapping patterns of neural activity simultaneously. Even though you've got one hundred billion nerve cells, you can't have two overlapping patterns. In other words, there is a bottleneck of attention. You can only allocate your attentional resources to one thing at a time.

Well when you look at a Page Three girl, the main information about her sinuous soft contours is conveyed by her outline. Her skin tone, hair colour after all is no different from anyone sitting here. It's irrelevant to her beauty as a nude. So in the realistic photo you have all this irrelevant information cluttering the picture and distracting your attention away from where it's needed critically - to her contours and outlines. By leaving all this out in a doodle or sketch the artist is saving your brain a lot of trouble. And this is especially true if the artist has also added some peak shifts to the outline to create an "ultra nude" or a "super nude".

What's the evidence for all this? Of course you can test it by doing brain imaging experiments comparing neural responses to outline sketches and caricatures versus full-colour photos. But there's also very striking neurological evidence from children with autism. Some of these children have what's called the savant syndrome. Even though they are retarded in many respects, they have one preserved island of extraordinary talent.