Patients with schizophrenia are able to correctly see through an illusion known as the 'hollow mask' illusion, probably because their brain disconnects 'what the eyes see' from what 'the brain thinks it is seeing', according to a joint UK and German study published in the journal NeuroImage.
The findings shed light on why cannabis users may also be less deceived by the illusion whilst on the drug.
People with schizophrenia, a mental illness affecting about one per cent of the population, are known to be immune to certain vision illusions. The latest study confirms that patients with schizophrenia are not fooled by the 'hollow mask' illusion, and that this may relate to a difference in the way two parts of their brains communicate with each other - the 'bottom-up' process of collecting incoming visual information from the eyes, and the 'top-down' process of interpreting this information.
Illusions occur when the brain interprets incoming sensory information on the basis of its context and a person's previous experience, so called top-down processing. Sometimes this process can mean that people's perception of an object is quite different to reality - a phenomenon often exploited by magicians. The new study, by scientists at the Hannover Medical School in Germany and UCL Institute of Cognitive Neuroscience in the UK, suggests that patients with schizophrenia rely considerably less on top-down processing during perception.
The study used a variation on the three-dimensional 'hollow mask' illusion. In this illusion, a hollow mask of a face (pointing inwards, or concave) appears as a normal face (pointing outwards, or convex). During the experiment, 3D normal faces and hollow faces were shown to patients with schizophrenia and control volunteers while they lay inside an fMRI brain scanner, which monitored their brain responses.
As expected, all 16 control volunteers perceived the hollow mask as a normal face - mis-categorising the illusion faces 99 percent of the time. By contrast, all 13 patients with schizophrenia could routinely distinguish between hollow and normal faces, with an average of only six percent mis-categorisation errors for illusion faces.