Cracking Visual Illusions: Why We See Imaginary Colors

A new scientific discovery has unravelled why we sometimes see colours that aren't there.

The phenomenon of 'colour afterimages' is when you see illusory - or false - colours after staring at real colours for a longer time. Through this, the brain can be tricked into seeing colour in a black and white image.

The cause of this illusion is the mechanism that allows us to see colours the same throughout the day, independently of light changes. Without it, the colour of the world would change as we are under yellow sunlight, a green canopy, or in a bluish shadow.

Scientists have long debated what causes colour afterimages, and how the brain creates them.

Dr Christoph Witzel , Associate Professor of Psychology at the University of Southampton, has found the missing link between the illusory colours we see and the neural mechanisms that produce them. The answer is in the cone cells in our eyes.

Dr Witzel said: "Colour afterimages is an age-old classical phenomenon, but the more we have understood about it, the more it has become the subject of confusion amongst experts and in textbooks.

"Some say the phenomenon is caused by responses in the light-sensitive cone photoreceptor cells in our eyes, while others have argued it's due to the neural pathways making us see opposing colours. Others have even suggested it's due to some kind of yet unknown mechanism in the brain.

"We've finally got a conclusive answer - colour afterimages are not opposing colours as everybody had thought. Instead, those illusory colours reflect precisely what happens in the cone photoreceptors."

Dr Witzel carried out extensive experiments to work out where in our visual system afterimages come from. He devised tailormade methods to measure the precise colours people see in afterimages and tested three different predictions.

In one experiment, 50 participants were asked to stare at a specific 'starter' colour and then immediately afterwards colour-match what they saw. In another experiment, each of 10 participants adjusted 360 times the colour of sustained afterimages in a special display.

Dr Witzel compared these measurements with computational models of different stages of neural processing: the photoreceptors, intermediate structures in the brain like the Thalamus, and the visual cortex in the brain.

"Across all the experiments, we found the same thing - afterimages are not caused by opposing colours, as many scientists have believed," he said. "Instead, they match what we'd expect if they were caused by how cone cells in the eye adapt to light.

"So, we have been able to say for certain that afterimages come from cone cells and not from other parts of the visual system."

He added: "The phenomenon of colour afterimages is not new, but this mass of data provides a full and coherent explanation for the first time. It is the missing link to fully explain what is happening in the eyes and the brain."

The research is published in the journal Communications Psychology .