The Fraser Spiral illusion (top) and Slant illusion

The visual illusions shown here challenge the way we look at the world. They are perplexing, deceptive and misleading, and scientists believe they can provide a profound insight into the software of the mind.

Computer scientists and psychologists are using computers to investigate how the brain reacts to visual illusions such as these. The results reveal new details of how our eyes and brain work together to generate a picture of our surroundings, and suggest new ways to fool our vision machinery.

“Visual illusions are some of nature’s most impressive magic tricks. They are deceptively simple and yet compelling — your brain is telling you lies,” says Prof Peter McOwan of Queen Mary, University of London, who has worked with Prof Alan Johnston of University College London.

The brain is the most powerful computer of all and exploring how it works is a “fantastic challenge”, he says. “Your brain is doing some amazing calculations as you read these words. Not only are you recognising the letters, the upright and top cross of the ‘T’, but you are also understanding what those letters mean when put together.”

Approximately half of the human brain, mostly the wrinkly surface at the back, is involved in processing the flood of information from the eyes as you read this sentence. Scientists are still arguing about the details of exactly what goes on in the brain — though we are gradually learning more through the use of brain scanners and studies of people with brain injury — but in rough outline, this is how we see: the optic nerves transmit signals containing visual information to the primary visual cortex, or V1, at the back of the brain, which breaks the image into small parts. Neural machinery analyses the patterns and orientation within these patches. We make sense of what we see as these signals move in two streams along networks of brain cells from the back of the head to areas involved in higher processing, although there are many waves washing information back and forth.

One stream of signals runs over the top of the brain, where extra processing is done by the V2 (which deals with the organisation of patterns) and V5 (which analyses motion), to figure out in broad outline where things are. The second stream passes information down the side of the brain, again via V2, to V4 — which seems to handle colour and is tuned to geometric shapes — and then on to a region called the inferior temporal lobe, which contains cells responsive to objects and faces. Thus, this stream shows what things are.

So the brain doesn’t simply reflect what is going on before us: it actively interprets the signals. It follows rules, based on experience, to build up a picture of its surroundings. But following these rules can sometimes lead it to the wrong conclusions.

One example is called the slant illusion. The mid-grey bars in these three pictures all have the same slope of 45 degrees. However, the brain’s perception of the slope is changed by the background, making it appear steeper in the middle image.

Scientists were able to design this illusion as their computer model of the way V1 works had predicted that it would be confounded by an image of a vertical black-and-white pattern meeting the grey diagonals.

In effect, because the model follows the brain’s rules for working out slopes, the computer is fooled by this illusion too. “Slopes, or spatial orientations, are an important part of the early visual calculations,” said Prof McOwan. “Slopes tell us something about an object — a rectangular table, for example, has straight edges — and slopes caused by perspective also tell us something about how far away an object is and how it's positioned in the scene.”

The brain cells in V1 respond to little parts of what we are looking at and rules are then used to turn these little chunks of what the eye perceives into the big picture. “When it comes to slopes, we know that in your brain you measure the orientation at each point in the scene at several different angles, rather than just at one or two,” said Prof McOwan. But the illusion results because “your brain is making a ‘mistake’ in its slope calculations”. The same breakdown of rules occurs in the Fraser Spiral Illusion. (The overlapping black arcs appear to form a spiral; yet they are in fact nested concentric circles.) For more illusions, see www.cs4fn.org/illusions.

Whatever the illusion, there is still much debate about why and how our brains are befuddled. “This is right at the frontier of brain research,” says Prof McOwan.