“Perceptions are not always, and very likely never, directly related to physical reality” Richard L Gregory

How is it that an arrangement of blobs is immediately recognisable as a pair of characters from a 17th century work of fiction?

Education and acculturation obviously play a part; anyone who had never heard of Don Quixote could not recognise the characters - or even if they knew of them, they might not be aware of Picasso’s acknowledgement of Daumier and a whole heritage of 19th century Illustration of the subject.

But would they see a thin man on a horse and a fat man on a donkey under the sun in a field of windmills? Because the horse is impossible, no sun ever shone like that and the windmills are ciphers, apparently taken from the symbols on a map.

The completion of images to make them recognisable is partly an automatic process. This can be tested quite simply in this illustration:

Close your right eye and focus the left on the square. Move toward and away from the screen until the circle disappears and the line appears continuous.

This process is not cognitive at all – the line will be continuous even when we know it is not, even when we understand the nature of the blind spot at the back of the eye  and know that our brain is ‘filling in’ it is simply not possible to stop it happening by an act of will.

Why is this so? My own view is that, with stone-age brains, living in an information age society, we are programmed by evolution to look for pattern and significance. Who has not experienced the hairs rising on the back of the neck from seeing something pass quickly by out of the corner of your eye? Was it really just a fluttering leaf or something more sinister? The ability to see the pattern of stripes lurking partly disguised in a stand of trees could mean the difference between survival and being eaten by a hidden tiger. As social omnivores, both predator and prey animals, recognising patterns could in our evolutionary past have been the difference between life and death, survival and starvation. And we have developed technological, social and artistic constructs that depend upon our need to see pattern.

The entire history of science and the scientific method exists to find patterns in the way the world around us works and to test those patterns against reality. Experimental methods such as the double blind trial have had to be developed to prevent subject and experimenter from seeing patterns where none exist. The need to find patterns manifests itself in spiritual manifestations, tealeaf reading, ley lines, all of which may actually have significance but are much more likely to be the result of coincidence; but it seems our default position is always to see significance in such phenomena.

The finding of strange, apparently supernatural, significance in random events is such a common experience that we even have a word for it – serendipity.

Of course, the picture of Don Quixote and Sancho Panza is not a Rorschach blot, a random collection of marks onto which the viewer projects an image of their own making. Picasso, like all artists, uses a subtle set of hints and signals which we, looking at the picture, complete to construct the image we ‘see’. In fact, of course, because perception is both an automatic and a cognitive process, biological and cultural, the further away in time we are from the creation of a work the more we will perceive it differently from the artist and his/her contemporaries. This is something that we simply have to take into account and live with, and is common to all creative work. A Byzantine icon, The Canterbury Tales, a symphony by William Boyce, will all have carried deeply moving significances for their audiences that are forever lost.