This article comes from the WeChat public account:Principle (ID: principle1687), author: Trafton, from the title figure: vision China

One

There are two gray dots on a background board with a gradient from black to light gray. Which one looks brighter?

The optical illusion of brightness contrast. | Image source: [1]

Although the two points are the same, they look very different in different backgrounds. This is a typical optical illusion of contrast in lightness.

According to the definition of “Psychology Terms”, Brightness(brightness) can be understood as a perception of the brightness of the surface of the object Subjective mentality. It is different from brightness (luminance), brightness generally refers to surface luminescence(or Reflective) strength.

When we look at an image, our brain can perceive a certain brightness at every position of the image. However, surprisingly, our perception of lightness is not always proportional to the amount of light emitted from the image area. On the contrary, our perception is the combined result of the actual color of the object and the amount of light shining on the object.

If you put a very dark cloth in a bright spotlight, the amount of light you get from it may be the same as the amount of light you get from a white sheet of paper in dim light, or even more. The brain always faces a challenge, that is, how to judge the light or darkness of a surface based on the energy it receives. In essence, the brain must calculate the result of multiplying two numbers, which is to combine the light level and the surface darkness to get the value of the input energy. This seems like an impossible taskBusiness, because an infinite number of pairs of numbers can produce the same result.

For more than 100 years, scientists studying the brain have been trying to find out the mechanism behind it. Some scientists believe that lightness estimation is an “advanced” process, including the 19th century German physicist Von Helmholtz(Hermann von Helmholtz), he was a pioneer in early vision research.

In this view, the brain estimates lightness based on a high-level understanding of the lighting conditions, shapes, and shadows in the environment it sees. Like many visual tasks (such as recognizing faces or objects), it may need to rely on our previous experience or on what we see Expectations.

A new study led by the Massachusetts Institute of Technology shows that lightness estimation may be a “low-level phenomenon”, which may not require advanced processing by the brain. It occurs before the visual information reaches the visual cortex of the brain, and may occur in the retina. These results are helpful for answering. The lightness estimation serves as the basis for many other visual analyses, and what mechanism constitutes this basic process.

Two

In the first set of experiments, neuroscientist Pawan Sinha and colleagues created an image of a cube. This cube looks like it is illuminated from the side, and one of the faces looks brighter than the other. In fact, this is a clever technique, the brighter surface actually has lower brightness. As early as more than 800 years ago, Chinese ceramic painters used this technique to create.

A schematic diagram of two graphs with differences in brightness. The left side looks more bright than the right side, but its brightness is lower. When the same gray point is on the point, the gray point on the left is more gray than the right It also looks brighter. In the experiment, two faces like this are placed on both sides of the cube.| Image source: reference source [2]

Researchers found that when the same gray dots are placed on both sides of the cube, the dots on the darker side actually appear darker. This is exactly the opposite of the standard brightness comparison results mentioned earlier. This result also suggests that lightness estimation may not be an advanced analysis that depends on lighting conditions.

The second set of experiments is to find where the lightness estimation process takes place. It is based on the strange fact: The image of our cognitive world is composed of the images of two eyes, and at the same time, we have almost completely lost the “original” information. We don’t know what the original image is and which eye it came from, we only know the merged view.

However, by using specially designed images and stereo glasses, the researchers found that the lightness estimation does not need to wait until the information of the two eyes is fused, and it has already appeared at that time.

That is to say, lightness estimation occurs very early, before the information of each eye is merged into a visual flow. The merger of vision occurs in a part of the cerebral cortex called V1, which is the first stage of visual processing in the cerebral cortex. This places a strict restriction on the processing position of the lightness estimation. The researchers suggest that important brightness calculations are most likely to occur on the retina.

Sinha said that the results of these two sets of studies mean that, if the lightness estimate is really a low-level process, and the pathway is located on the retina, then this may be an innate mechanism. This is what the vision system is ready to do from birth.

Three

Researchers explore this hypothesis by studying blind children who have recently recovered vision. Sinha launched a project called Prakash in India, whose mission is to treat children with preventable blindness such as congenital cataracts. Many children receiving treatment continue to participate in scientific research on visual development, although treatment does not depend on participation in the research or not.

If lightness estimation is really a natural mechanism, then children with congenital blindness should have the illusion of lightness contrast after their vision is restored, which is what the researchers foundof. Nine children between the ages of 8 and 17 underwent cataract surgery and were tested within 24 to 48 hours of removing the bandage. When their eyesight began to recover after the operation, they would soon be affected by the illusion of lightness contrast. This provides further evidence that lightness estimation may be based on simple neural pathways and does not require any prior visual experience.

In a 2015 study, Sinha found that children who have just returned to normal vision will soon be affected by two other optical illusions, namely Ponzo illusion and Miller-Lyer illusion< /strong> These two illusions involve judging the length of line segments based on visual cues.

Ponzo Illusion (left) and Miller-Lyer Illusion (right). What Ponzo illusion says is that there is a pair of straight lines converging to a certain point, in which two horizontal lines of the same length are drawn. In this case, we visually think that the upper horizontal line is longer. The Miller-Lyer illusion refers to two parallel lines of equal length, and the line with arrows pointing at both ends looks longer than the line with arrows pointing at both ends.

“The conclusions drawn from this study also seem to be consistent with the conclusions drawn from the lightness study. That is to say, We attribute many phenomena to advanced reasoning processes, but in reality, they may be behind There are some very simple pathway mechanisms born in the brain.” Sinha said, “These results help us understand how the nervous system solves the complex challenges of perception and understanding of the surrounding world.”


Reference source:

[1]http://news.mit.edu/2020/study-visual-illusion-brightness-0617

[2]https://www.sciencedirect.com/science/article/abs/pii/S0042698920300730

This article comes from the WeChat public account:Principle (ID: principleia1687) , Author: Trafton