RSS Feeds

Email a Friend Printer Friendly

Optical illusions, mirages that don't deceive

July 24, 2006

The aim of this paper is to dispel the excessively widespread myth that optical illusions are errors of the visual system. In 1978, Stanley Coren and Joan Stern Girgus published one of the most significant works of scientific literature in the last few decades, entitled "Seeing is Deceiving: The Psychology of Visual Illusions". It is difficult to digest that someone who is seriously dedicated to the study of vision affirms that vision deceives us. If this were true, we would not be able to trust what we read in the book, since we would read it using our vision. We can therefore deduce that this is nothing more than a strategy to get us to read it.

If we take a look at some optical illusions, we realise that this type of visual configuration can often be found in situations we meet every day. In fact, these visual configurations help us in our environment. This can be seen in the famous checkerboard proposed by EH Adelson in 1995 (Adelson, 2000), a well-known optical illusion (checkershadow illusion). The illusion can be seen in the picture on the left. You will probably have no difficulties in stating that square A is much darker than the centre square, labelled B. However, in the picture on the right, the colours of the two squares appear to be very similar. In fact, if we measure the illuminance using a photometer, we find that the two squares have exactly the same illuminance.

So why does square A look darker?
The visual system (VS) must determine the colours of objects we encounter. In the example of the checkerboard, the VS must identify the colours of the squares. If the VS relied only upon measuring the illuminance of the squares, we would perceive squares A and B as identical. However, this would not allow us to interpret the "real" situation of the checkerboard correctly, that is, that the centre square is white and the B square is black. Although with regard to physics square A and square B have the same illuminance, this is neither relevant nor useful with regard to perception, since the equal illuminance does not allow us to identify the different colours on the checkerboard.

The VS uses various "tricks" to compensate the effects of shadows, maintaining the continuity of the colours (perceptively, one does not stop seeing a certain colour when the surrounding light fades, even though, from the perspective of physics, the colour of a surface changes according to the light it receives). The VS is not a good measure of "physical data", but this is not its purpose. The main role of the VS is to decipher the information in the image and thus perceive the nature of the objects seen.

From this perspective, perception must be understood as a procedure for processing the information in the surrounding environment rather than a reflex of basic physics. Perception, therefore, involves a series of operations to transform, analyse, synthesise and activate knowledge. Optical illusions generally use these operations, therefore it is unfair to say they "deceive" the system. On the contrary, we can assume that the VS, with its ability to perceive in a different way to what is shown by physical measurement, protects us rather than deceives us.

We aim to demonstrate that, in general, optical illusions do not deceive us in our everyday interaction with our surroundings but rather they give us more appropriate knowledge to direct our behaviour. In short, it seems that it is to our advantage, and even necessary, that we "suffer" some of these optical illusions.

Universitat Autonoma de Barcelona