Prof. Shaul Hochstein from the Hebrew University studies learning and memory processes of visual information. He and his colleagues claim, contrary to conventional thinking, that the speed with which a person recognizes objects and the extent to which he remembers them depends on their importance to his survival and not on their complexity
Uri Nitzan
Three examples from visual information processing experiments.
In picture A, the subject needs a tenth of a second,
On average, to pick the outlier in terms of color or
the bias; In picture B, it takes an average of a second to
Choose the unusual - the line with a different slant
The rest of the lines have the same color. in picture C
It takes a tenth of a second to locate the face, which it is
A relatively complex bone
Building an image of the surrounding world is one of the most complicated tasks the brain performs, and scientific research is only beginning to decipher the mechanisms responsible for its execution. The visual system is required to process the image of the world and build an internal representation of it. Only after processing the image, and discovering what is in it, can the person respond to the events that are happening around him.
Professor Shaul Hochstein from the Interdisciplinary Center for Neural Computing at the Hebrew University investigates processes of visual learning and memory, which participate in the construction of the image of the world. These processes are based on real-time changes in the neurons in the brain. "The general issue that concerns us is the limits of the brain's ability to change," says Hochstein. "We are trying to answer two central questions: How do humans learn through the visual system? And where do the processes of visual learning and memory storage take place? to answer to
We investigate these complicated questions in the visual system at several levels and with several methods: electrical recording from individual neurons, brain imaging of entire areas of the brain, and behavioral tests that test visual learning and memory abilities, which allow us to describe the laws that shape the image of the world."
The initial visual stimulus consists of millions of photons of light, which fall on each of the hundred million retinal cells in the eye. In order to decipher and identify what is "seen", the brain analyzes different aspects of the image in a processing process, which involves more than thirty vision centers in different areas of the brain. The vision centers are organized in a hierarchy, from low centers where the initial processing of the information coming from the eye is carried out, to high vision centers that give the representation of the image abstract characteristics such as meaning and context. The higher vision centers are also responsible for identifying objects and perceiving categories of animals, houses, utensils, food, faces, and the like.
In one of the lower processing stages, the brain recognizes the boundaries between lighted and shadowed areas, thereby giving the visible image a primary meaning related to the boundaries between different objects or parts of objects. In the centers the activity of the brain cells is increasing and already represents outlines (lines with an angular inclination, in a certain place and thickness); From these lines a representation of geometric shapes is built, to which color and movement are added. At the end of the process, a representation of complex objects such as a face is obtained.
The evidentiary information is processed in two main and complementary ways. The first way
It is "simultaneous processing", which occurs when the visual system simultaneously processes aspects
different of the image. In this form of processing only at the end of the process do the colors merge together,
The movement, the shape, the location in space and other variables, into a complete picture. the road
The second is called the "serial processing". In this type of processing, the information passes from a visual center
One to the one that follows, when each center "takes in information that comes to it from centers
lower, upgrades it and passes it on to higher vision centers
More," explains Hochstein. "When a person reads a newspaper, for example, the visual centers build up
lines from the black dots that are printed on the page, from the lines the letters are composed,
From letters words, and from words meaning".
Classical visual learning theory assumes that it is easier for us to recognize and remember a shape
simple or a particular feature of an image, than a complex shape. For example, in the picture
A. We recognize in about a tenth of a second the unusual, whether it is a color
The exception of the line and whether it is its different angular tilt in space. "Vs
However," explains Hochstein, "identifying the anomaly in picture B is a more difficult task.
Here the picture is more complex, and the visual system must simultaneously process related information
for color and information related to angular bias, and combine the results of these renderings for
Each line separately. In this case the observer needed an average of a full second to find out
You are the exception."
According to the classical theory, the detection time of an object increases in direct proportion to its complexity
And the number of objects in the field of vision within which the viewer is required to isolate himself. Hochstein
and his colleague Dr. Merav Ahisher, also from the Center for Neural Computation, developed over the years
The latter is a theory that redefines the visual learning and memory processes,
and offers an alternative to the principle of complexity. "We believe that the detection speed of
Objects, and their recall, do not depend on the simplicity of the object but rather on the degree of importance
and its relevance to our lives," explains Hochstein. "The brain uses both processing
Simultaneously and in serial processing, but the information that enters awareness and is stored in memory arrives
First and foremost from the higher vision centers. The final product of the processing is this
which penetrates our awareness first".
When the researchers asked subjects to look at a group of objects like the one in picture C
And to discover the face - an essential action for the survival of humans - they performed the
The task in a tenth of a second, a speed similar to that required to detect color or
Unusual tilt angle. "When we asked them to discover a less 'important' object like a tool
work, they needed a longer time," said Hochstein. It is, though
that the complexity of human faces is much higher than that of other objects.
According to Hochstein, "The working assumption we formulated is that the brain locates, recognizes and remembers
'The main thing', and erases and forgets the therapist who is in his environment. In the test we conducted we found
that the subjects fail to remember and identify additional objects that are arranged around the face
who discovered; They also did not remember the exact details of the faces themselves. the identification
The rapidity of the main involves a lack of awareness of activity in lower regions of
Vision system".
An important pillar in establishing the new theory was the study of the physiology of the visual system
and the properties of the nerve cells at the different levels in the hierarchy. In many studies on system
The vision recorded the electrical activity arising in these cells in the brains of animals
laboratory, in response to the presentation of simple or complex images. It is possible, for example, to teach
Monkeys recognize a certain shape, such as a face, and touch it with their finger when it is shown on
on the computer screen. Recording the electrical activity during the study, or in the process
Identifying the bone image allowed the researchers to locate the visual centers involved in learning
and in object recognition.
"The electrodes that researchers place in the brain do not stimulate nerve cells", he emphasizes
Hochstein, "but record the natural electrical activity. The monkeys don't feel
This is because there are no pain receptors in the brain, and the high level of safety of the method
Allows doctors to perform a similar electrical recording also in patients with syndromes
Certain neurology".
In the next step, the level of compatibility between the findings in monkeys and Benny's visual system was tested
A person. "We conducted fMRI brain imaging tests on the subjects while they observed
in pictures of various objects on a computer screen. In the test you can see the blood supply
to the different areas of the brain, and the more active an area, the stronger the blood flow to it
More. Using an fMRI test, it was proven that there is a correlation between the location of the visual centers of
The human, which are active during face recognition, to locate the same centers in the monkey.
Hochstein said. This activity was detected in several regions of the posterior-lateral lobe of
the brain. These are areas that the damage to them caused in some cases the loss of the ability to recognize
faces A famous example of this type of injury is described in the neurologist's book
The American Oliver Sachs, "the man who thought his wife was a hat".
https://www.hayadan.org.il/BuildaGate4/general2/data_card.php?Cat=~~~442901389~~~174&SiteName=hayadan
One response
There is a theory that says that symmetry of the body and face looks more beautiful to us because it is easier for us to perceive with our senses (and we also do it faster) such geometric shapes and as for less symmetrical shapes, we will recognize beautiful faces in the crowd faster than less attractive faces. If you connect this to the theory in this article "the speed with which a person recognizes objects and the degree to which he remembers them depends on their importance to his survival and not on their complexity" this means that attractive people are more important to our survival than less attractive people, which of course is not necessarily true.