When perceiving illusory movement when considering the Pinna-Brelstaff optical illusion, which causes static circles to move in different directions when moving the head towards or away from the image, there is a small (15 milliseconds) delay in the work of the brain regions that are responsible for the perception of movement. This is by means of an experiment on macaques, in whose brains electrodes were implanted, Chinese scientists have shown. Article published in The Journal of Neuroscience.
Optical illusions (if you have a poor idea of what it is, you can take our New Year's test "Duck or bunny?") Are poorly understood in matters of the visual system. Ten years ago, scientists showed that when viewing optical illusions of movement (for example, the same Pinna-Brelstaff illusion), the observed illusory movement appears to the brain as real: in its perception, the middle temporal region is active, which is responsible for processing real movement.
At the same time, it is still not clear how exactly the illusion of movement arises in a static picture. Chinese scientists decided to study this issue in more detail under the leadership of Junxiang Luo from the Institute of Neurosciences of the Chinese Academy of Sciences. They decided to focus on activity in two areas of the visual cortex: the dorsal mid-upper temporal area and the mid-temporal area.
In their experiment, macaques took part, in the visual cortex of which electrodes were implanted. Before starting the experiment on macaques, scientists conducted a study involving humans. Nine volunteers were shown Pinna-Brelstaff illusions composed of Gabor spots arranged in several circles. Depending on the illusion the scientists were trying to reproduce, the spots were located in circles with an inclination of 45 degrees to the right, 45 degrees to the left and straight ahead, and to facilitate the task, the circles either expanded or narrowed. Participants had to be told how the circles move, depending on whether the drawing gets further or closer. In the reverse task, the circles were not static, but moved counterclockwise and clockwise: in this case, the participants had to note whether they saw an expansion or a narrowing of the circles.
Experimental paradigm and used illusions created from Gabor spots.
Depending on the inclination of the Gabor spots in the Pinna - Brelstaff illusion, the direction of movement seemed different to the participants: for example, when the spots were tilted to the left, the circles illusoryly moved clockwise when expanding and counterclockwise when narrowing (and the opposite pattern was observed when the spots were inclined to the right). At the same time, the real counterclockwise movement of the spots created the illusion of narrowing for spots inclined to the left in the circles, and clockwise movement created the illusion of expansion (again, the opposite pattern was observed for spots inclined to the right). At the same time, the circles consisting of directly oriented Gabor spots do not cause any illusion of movement.
The pattern of observed movement under different stimulation.
Then the scientists conducted an experiment with the participation of two macaques, which initially learned to recognize different directions of movement in circles with directly oriented Gabor spots: for this, the circles were twisted clockwise and counterclockwise, and also narrowed and expanded, after which they were asked to select the macaques on the computer which movement. they watched. After successful training, the scientists repeated the experiment conducted on humans on monkeys, changing the speed of movement and doubling the maximum speed. Scientists have found that macaques perceive illusory movement in a similar way: the threshold for perceiving a stimulus depends on the speed of rotation or expansion / contraction.
Promotional video:
Observed direction of movement with different direction of real in macaques.
After scientists were convinced that macaques see the Pinna-Brelstaff illusion the same way humans see it, they conducted a final study of the activity of the visual cortex of their brain. Scientists have found that neurons in the dorsal part of the middle upper temporal region and the middle temporal region are equally activated both when observing real movement and when observing illusory movement (both in a circle and expansion / contraction). At the same time, when observing the illusory movement, the neurons of the dorsal part of the middle superior temporal gyrus are activated 15 milliseconds later than when observing the real movement.
Beginning of activation of neurons in the middle temporal region and the dorsal part of the middle upper temporal region when observing real (above) and illusory (below) movement.
Both of the studied parts of the visual cortex are responsible for the perception of a complex movement - for example, the one that is observed when the circles rotate as they approach or move away. In this case, the dorsal part of the middle upper temporal area is activated earlier, apparently delimiting the nature of the observed movement for further processing by the middle temporal gyrus. When observing the illusory movement (the one that occurs in optical illusions), the neurons in this area, according to scientists, need additional processing time. Based on the fact that the perception of illusory movement in the Pinna-Brelstaff optical illusion turned out to be similar in macaques and humans, it can also be assumed that a similar delay can be observed in the work of the visual cortex of the human brain.
While cognitive and neuroscientists are studying how people perceive optical illusions, developers are teaching neural networks to create them themselves. They do it, however, not very well.
Elizaveta Ivtushok
