How Our Brains Set the World Spinning

By Carl Zimmer | April 24, 2012 5:00 pm

If there’s ever excuse to publish an optical illusion as cool as the “Rotating Snakes,” I’ll take it. This illusion was invented in 2003 by Akiyoshi Kitaoka of Ritsumeikan University in Japan, and ever since, Kitaoka and other scientists have been trying to figure out why it works. A new paper by Stephen Macknik at the Barrow Neurological Institute in Phoenix may have the answer.

As you’ll notice, the circles seem to rotate in response to where you look at the illusion. So Macknik and his colleagues tracked the movement of people’s eyes as they gazed at two of these wheels on a computer screen. Their subjects kept a finger pressed on a button, lifting it whenever they seemed to see the wheels move.

Macnick and his colleagues found a tight correlation between the onset of the illusion and a kind of involuntary movement our eyes make, known as microsaccades. Even when we’re staring at a still object, our eyes keep darting around. These movements, called microsaccades, help us compensate for a peculiar property of the eye: if we stare at an object for too long, the signals each photoreceptor sends to the brain become weaker. Microsaccades refresh the photoreceptors with a different input and breath new life into our perception.

Unfortunately, the jumps of our eyes get in the way of our perception of motion. If we see a snake slithering along in a desert, we don’t have to register an entire image of the snake at one instant, then another image at the next instant, and then compare the location of the two images, in order to figure out that the snake is on the move and we might want to jump out of the way. Instead, we only have to sense rapidly changing light patterns in neighboring parts of the eyes. If certain neurons in the vision-processing regions of the brain gets a sudden, strong signal from the eye, they register motion.

Normally, our eyes can register motion despite the fact that they are also performing microsaccades. Our brains can tell the difference between a shift brought on by the movement of an object and one brought on by the movement of our own eyes. But thanks to the strong contrasts and shapes in the Rotating Snakes Illusion, we get mixed up. Our motion sensors switch on, and the snakes start to slither.

Reference: “Microsaccades and Blinks Trigger Illusory Rotation in the ‘Rotating Snakes” Illusion.’ Otero-Millan et al, The Journal of Neuroscience, April 25, 2012 • 32(17):6043– 6051

[Image: From Akitaoka’s snake web site. Many more big screen versions there!]

[Update: I revised this post to correct the explanation of microsaccades and their function. Thanks to John Kubie for his comments and follow-up emails.]


Comments (27)

  1. Cool. But why do I (and I think most people) see motion only in the periphery?

    Whichever circle I look at, that’s stationary and the other circles are all moving.

    I’m guessing that the fovea (the part of the retina which codes the point of fixation) is better at compensating for microsaccades than the parafovea.

    Guessing further, perhaps the parafovea doesn’t normally need to bother compensating for microsaccades because the retinal “image” is more blurred in the parafovea. Something about the illusion (the high contrast perhaps) overcomes that, giving the motion sensation.

  2. What about something as simple as Extraocular Muscle fatigue (Instability?)

  3. Very interesting. But if these microscaccades cause the illusion, I wonder why we dont get any similar illusion of movement when we see just a rectangular object, with contrastring stripes of colours arranged all along it from left to right.

  4. Mephane

    @Jon Brock: I have the same experience. Any circle not currently in my focus appears in motion. And it even works if they are cut off by the edge of the browser window.

  5. It would be interesting to know whether someone who has had his sight restored after a long period of blindness experiences the illusion.

  6. Mikalai

    I just did an experiment.
    Blinking with different frequency.
    The result is next : for fast (60 Hz) blinking the rotation is present,
    for lower 10 Hz frequency the rotation stops. Conclusion : there is a stroboscopic effect, that means we also have to include in analysis the speed of information processing.
    Of course it is not statistically enough to build the conclusion one one person, but …

  7. John Kubie

    I don’t see how this has anything to do with “the quirky evolution of backwards retinas”. Please explain.

  8. John–The linked paper shows how the blind spot is a result of the way our retinas are pointed backwards, with the optic nerve going through a portion of it. It’s my understanding that saccades allow us to compensate for that.

  9. Jenn

    Does anyone else not see the rotation when they close one eye? Is that normal, or the result of my migraine?

  10. gaddeswarup

    I tried several times but do not see any rotation or snakes. does age has something to with it? I am seventy.

  11. Burkhart

    There are quite a number of optical illusions thAt disappear as one stops any saccades, but fixates.
    Most of these illusions are geometrical illusions. Interested?
    Read “Illusory Illusions: The significance of fixation on the perception of geometrical illusions” by
    Fischer,B; daPos,O; Stürzel,F. (2003)Perception 32, 1001-1008.

  12. A colleague and I wrote one of the papers that was cited in this new paper by Otero-Millan, Macknik, and Martinez-Conde (

    Some other fun facts about this illusion are:

    Color makes the illusion stronger. There is also illusory motion in gray-scale versions of the image, but there is no set of four grayscale values that creates as much illusory motion as the colored version that repeats white, yellow, black, blue (and always moves in that direction). Kitaoka has also used red and green, and other colors, besides yellow and blue.

    If you keep your eyes still, the motion will slowly grind to a halt over the course of about 4-6 seconds. One cannot keep the eyes completely still (microsaccades will occur, and as the current paper shows, they will cause the illusion to start up again) but you should be able to keep your eyes still enough to see the motion die down or stop completely. This suggests there is something that continues to change dynamically within the visual system, over the course of 4-6 sec, that drives the illusory motion–and then you have to move your eyes to start the process over again if you want to see more motion.

    Not everyone sees the illusory motion. My mother doesn’t. There are always a few students in my classes who don’t see it. So either their eye movements are different, which now seems unlikely given this new paper, or else the dynamic change that occurs over 4-6 sec is different for these people.

  13. Carol

    First of all, thank you for such fascinating exercise! Peripherally, I see three of the snakes in motion as I’m fixated on the fourth; however, as I blink and refocus on the middle circle, I see all four snakes in motion and the center circle slowly in motion for about three seconds and it stops. At the end of my 5 seconds of staring into the center my eyes do get tired and need to blink again…fun!

  14. I am legally blind in my right eye, due to the advent of the “dry” form of AMD (Age-related Macular Degeneration). Although I can still see lawyers, the “Rotating Snakes” illusion does not work for me. Has anyone else with this form of AMD noticed this effect ? I am also incapable of viewing a 3D movie, but understand that this situation should not trouble me.

  15. Marshall

    Mikalai: there is no way you can blink your eyes 60 times in a single second. I’d be surprised if you could even do 6.

  16. John

    Sitting experimenting, I found
    1. relaxing my eyes, looking at an imaginary horizon, the movement stopped.
    2. Gazing at the horizon through the centre point of the centre, the 4 outer circles seemed more 3 dimensional, Each outer circle became a spiral with the centre furthest away.
    Can this be used to test for sight defects or perhaps any other (neurological) abnormalities.
    I do not completely understand the explanation of ‘why’ it works but feel there may be some practical implications of the results.
    I am fairly short sighted with one eye worse than the other and looked atthe picture without glasses. It did not seem to make any difference which circle I looked at, the other 3 moved.

  17. brooke

    U can make the snakes stop if you stare hard enough but they slither away, the second u pull ur focus away.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

The Loom

A blog about life, past and future. Written by DISCOVER contributing editor and columnist Carl Zimmer.

About Carl Zimmer

Carl Zimmer writes about science regularly for The New York Times and magazines such as DISCOVER, which also hosts his blog, The LoomHe is the author of 12 books, the most recent of which is Science Ink: Tattoos of the Science Obsessed.


See More

Collapse bottom bar