Optical illusions are intriguing and sometimes baffling. In the 1990s “Magic Eye” pictures were everywhere, from mall kiosks to the back of cereal boxes. The “Magic Eye” were pictures that were distorted in such a way that a 3D image would over time appear to the viewer. The image never appeared to me, no matter how long I started at it—both with and without glasses. Another optical illusion is the famous “My Wife and My Mother-in-Law” illustration created by W.E. Hill in 1915. Depending on your perception, you either saw a young woman or an old woman. Whenever I saw this illustration, I managed to see both images.
Still, what causes optical illusions in the first place? As to what causes optical illusions, the eye is constructed so that the cones and rods are placed on and around the retina. The cones detect color and the rods detect low-light contrast. They convert light into signals that are carried to the brain. On the edges of the retina there are fewer cones and rods, whereas the center has more. Seeing things from the corner of the eye can create deceptive images. Also, exposure to alternating patterns, brightness or certain colors can affect our perception, leading to an optical illusion.
Being that many things can lead to optical illusions, how does the brain differentiate between the foreground and the background when it comes to viewing an object. For example, in the famous “Rubin’s vase” optical illusion, you either see a vase or two faces who look like they are about to kiss each other.
Researchers in the lab of Professor John Reynolds at the Salk Institute looked into this and what they found could lead to better understanding of conditions where perception is disrupted, such as schizophrenia. When viewing a scene, individual neurons in the brain’s cortex receive information about a small region of the scene. Neurons that receive information from the border of an object don’t have much context about which part is the foreground.
Scientists previously found a group of cells that quickly signal which side of the border belongs to the object. Researchers hypothesized that as information from the eye goes into the brain, additional calculations occur until the brain build a model of the scene. This is known as the “feedforward” pathway. Other scientists hypothesized the importance of the “feedback” pathway. This is where the downstream areas of the brain must first process information, then send clues to neurons in upstream areas to assist in figuring out the where the border is located.
Researchers in Reynolds’ lab wanted to see which hypothesis is correct. They used electrodes to record activity of neurons in different layers of an animal’s brain cortex as it viewed a picture of a square object on a blank background. The scientist first learned which neurons were processing information from a part of the border that marks off the square and the background. Next, they measured the timing of the border ownership signals in these neurons and compared this to neurons in different layers.
What they found is that that the earliest signals on border ownership occur neurons in the deep layers of the brain’s cortex. This finding supports the importance of the feedback pathway for interpreting the border, since the feedback connections arrive at it and leave from neurons from the deep layers of the brain. Researchers also found that the neurons stacked vertically in different layers in the cortex shared the same preference for border ownership. This suggests that feedback is organized in a systematic way.
“As we come to understand the architecture of the brain and how ensembles of neurons communicate with each other to build up our internal representation of the external world, we are better positioned to develop diagnostic tools and treatments for brain disorders in which these internal representations are distorted, such as schizophrenia,” said Senior Postdoctoral Fellow Tom Franken, who worked with Professor Reynolds on this research. “The hallucinations and delusions associated with schizophrenia may be associated with the disruptions of feedforward-feedback loops.”