Dec 27 2010

The Spinning Girl Illusion Revisited

Three years ago I wrote a post about a popular illusion – the spinning girl or silhouette illusion. This is a popular online illusion, and also remains my most popular post. (Original illusion by Nobuyuki Kayahara here.) The popularity of this illusion seems to be tied to the fact that it is used in many online quizzes, with the claim that the direction in which you see the girl spin will tell you which side of your brain is dominant. In my prior post I primarily addressed that claim – explaining that the “left brain – right brain” thing is all nonsense, and which way the girl appears to spin tells you nothing about your personality or talents. (Briefly – while many neurological functions are lateralized to one side of the brain or the other, both hemispheres are massively connected and work together to form your abilities and personality.)

The real question prompted by this illusion is why do we perceive it as rotating one way or the other, and is there a preference. It turns out, most people will see the girl spinning clockwise. You can get her to switch and spin the opposite way to your original perception – but when first looking at the illusion most people will see her spinning clockwise.

Psychology professors Nikolaus F Troje and Matthew McAdam wrote a paper explaining this preference, and not surprisingly it has nothing to do with hemispheric dominance. It is purely a visual phenomenon – they write:

Here, we show that this rotational bias is in fact due to the visual system’s preference for viewpoints from above rather than from below.

Our visual system has many such biases and preferences. In effect, our brains process visual information with many default assumptions that are true most of the time. Many optical illusions are based upon creating a special situation in which one or more of these assumptions are false. For example, our visual system has a bias for lighting from above, assumes that smaller objects are farther away, and assumes that if one object overlaps another it must be relatively closer.

Troje and McAdam did some experiments with 24 subjects, playing with the apparent camera angle of the image. They found that there does not appear to be any rotational bias (preference for clockwise or counter-clockwise). The only bias they documented was the viewed from above (VFA) bias.

They also point out that silhouettes are inherently ambiguous – they do not provide clues for how to construct a 3-D image from the 2-D information. However, in this illusion there are subtle and conflicting clues for direction. They write:

Tracing the end of the outstretched hand shows that the camera elevation with respect to the hand is only 6.0 deg, which is a little less than the value for the foot. That means that the figure was rendered with a perspective camera positioned about 75 meters away from the figure (assuming a vertical distance of 1mbetween foot and hand). According to this perspective cue, the view from above and therefore the clockwise rotation is the ‘true’ rotation. However, another cue to the rotation of the figure is provided by the shadow the feet cast on the ground. The ellipse circumscribed by the shadow of the outstretched foot—at least assuming that the ground is horizontal and that we are looking at it from above—clearly suggests counter-clockwise rotation.

Their experiment and explanation is elegant, but it turns out the answer to this illusion was already available online. In addition to my post debunking the “left brain-right brain” claim, psychologist Michael Bach had also explained in 2007 the viewed from above bias as the true explanation. While I appreciate the extra detail and data provided by the new study, the answer was available online all along. (It should be noted that Troje and McAdam reference Bach in their paper.) Yet interestingly Bach’s explanation did not get much attention, while this new paper is being credited as finally debunking the silhouette illusion as personality test.

It is good news that if you search on the various relevant search terms for the illusion, my blog post ranks highly (it is the first hit when you search on “the spinning girl”, and now the Troje and McAdam paper dominate the rankings as well. Hopefully this was all a good opportunity to teach the public a little bit about neuroscience and visual perception, and to make a dent in the popularity of the left brain-right brain myth that is so common.

This is also another episode that points out the frustrating disconnect between pop-psychology and actual psychology. So much of what passes for psychology in the public domain is nonsense, or is simply false or outdated information. Yet there are many interesting and useful psychological experiments out there that the public is not aware of. Psychologist Richard Wiseman discusses this briefly on the SGU, and says this was the motivation for his book, 59 Seconds, in which he debunks a lot of pop-psychology but then gives useful knowledge from the actual research.

I don’t know if the internet and “web 2.0” is helping or hurting this phenomenon. At least for those who are interested, the real information is usually available, and often long before it finds its way into the official literature. In this case it took three years to catch up – but better late than never.

20 responses so far

20 thoughts on “The Spinning Girl Illusion Revisited”

  1. jimspice says:

    To complicate the matter of perspective, is it not also subjective as to what constitutes clockwise and counter-clockwise? Are we imagining ourselves standing on the clock looking down, or are we looking looking up at a clock with the person behind it. It would seem to complicate the viewpoint preference. Two subjects reporting conflicting rotations may, in fact, be seeing the same thing.

  2. medmonkey says:

    I can’t get the girl to go clockwise, even if I cover her shadow! This is really bothersome, especially since I feel like I used to be able make her switch 🙁

  3. jbc says:

    I think there’s an interesting neurological component to the illusion, but I don’t mean any pop-psychological stuff about left brain vs. right. I think part of the popularity of this version of the illusion might be that it provides an excuse for a (mostly heterosexual, mostly male) readership to stare at the pirouetting outline of a naked, attractively proportioned young woman. So, while the intellect is engaged with the interesting perceptual problem of flipping the spin direction, and the moralizing impulse is effectively stymied by the excuse that “hey, we’re doing science“, the hard-wired reproductive impulse is being stimulated.

    It would be interesting to do some sort of A-B test using a less-sexually-provocative silhouette, and measure the time spent staring at the spinning image. I hypothesize that straight men would spend significantly more time staring at this image than they would at the desexualized alternative, while straight women and gay men would not (or maybe they would, but to a lesser degree than the straight men?) I’d be curious, also, what effect would be shown by lesbians, and how it would compare to that of the straight men.

  4. tmac57 says:

    I was able to change back and forth by saying the opposite of what I was perceiving.For instance when she appeared to be facing the rear,I said “forward” every time, to counter my perception.It may take a minute or two.

  5. cwfong says:

    One reason we tend to first see the dancer spin off her right foot is that ballet dancers (and ice skaters) most often do that. So we expect to see that first in the illusion. The problem then is that we find no further clues that will confirm our expectations.

  6. wfr says:

    Everyone who uses the terms “clockwise” and “counterclockwise” without stating the assumption “as viewed from above” is obviously top-brained.

  7. Khym Chanur says:

    If I focus on her head I can get her to move counter-clockwise. If I concentrate elsewhere… well, it used to be that she went clockwise, but now she goes back and forth. Weird.

  8. elmer mccurdy says:

    I couldn’t get her to switch last time I saw this but now I can. The trick for me is to follow her food and adjust my eye focus closer or farther away as appropriate as she moves.

  9. elmer mccurdy says:

    No, I didn’t really mean “food.” Speaking of which, the microwave just rang.

  10. HHC says:

    She changes readily, clockwise or counterclockwise depending on whether I focus on the upper or lower torso. The silhouette is actually bad form for a dancer or an ice skater. Why would you want to stare at this thing? Its much more fun cognitively when you complete spins in either direction on the dance floor or ice.

  11. elmer mccurdy says:

    I’d be curious to see if there are different tendencies among people who are in the habit of reading from right to left.

  12. elmer mccurdy says:

    Oh, I just read the post a little more carefully. Never mind.

    Anyway, read dancing is more fun if there are real naked people.

  13. tmac57 says:

    I agree with Elmer.I can now readily switch directions if I follow the extended foot,and then imagine the toe circumscribing a circle in the opposite direction.

  14. BGH122 says:

    JBC, I didn’t even notice she was naked until you pointed it out. Maybe I’m not as heterosexual or male as I thought.

    Back when I studied psychology, psychology of perception was easily the most interesting module and illusions, especially their worth (or lack thereof) in studying perceptual phenomenon, were of great interest to me too. I think the most interesting thing that perceptual illusions do is show us how our underlying visual heuristics work, as Novella touched upon in the article. Beyond that they’re pretty worthless since they’re just too divorced from real world perceptual experience to be ecologically valid for anything. Sadly these ‘test your IQ by seeing how many circles rotate!’ illusion-based hoaxes are endemic online.

  15. BillyJoe7 says:


    That is barely believable.

  16. BillyJoe7 says:

    …but seriously, that is exactly the first thing I noticed. Thereafter I spent some time watching the different parts of her anatomy spinning around. Eventually I think I did pay some attention to the point of the exercise though I’ve entirely forgotten what that was.

  17. tmac57 says:

    Although this is entirely obvious,I just now realized that the extended leg is the left, when she is perceived to be spinning counter-clockwise,and it is the right leg when spinning clockwise.

  18. TheRedQueen says:

    here is a related TED talk featuring Beau Lotto –at about minute nine in the talk, he demonstrates a spinning illusion that changes direction every time a person blinks

    With the spinning dancer illusion above, my hack is to look at the shadow foot when I want the direction to change from clockwise to counter-clockwise.

    Good of Dr. Novella to explicate that this illusion has no utility in determining brain hemisphere dominance 🙂

  19. ccbowers says:

    One problem that I see with attempting to look this question of whether people see the woman rotating “clockwise” versus “counterclockwise” is that there are many factors that would affect this preference, and some of these are reasonable and rational (and separate from the visual aspect).

    Most people are right handed (or footed in this case) and that lends itself to a rotation bias. This is reflected in the sport in which we see rotation the most, figure skating. “Clockwise” and “conterclockwise” rotations are not equally likely in real life, and evaluating this as a purely visual bias would have to take this into account.

  20. cwfong says:

    Ballet dancers (if not skaters) for the most part spin clockwise throwing their right foot out and around with the left foot as the fulcrum.

    We expect to see that first in the illusion. The problem is we find nothing in the illusion that will confirm our expectations. The brain then searches for clues by switching things around, at least for a while.

Leave a Reply