The great Renaissance scholar and artist Leonardo da Vinci left a legacy of paintings that combined beauty and aesthetic delight with unparalleled realism. He took great pride in his work but also recognized that canvas could never convey a sense of motion or of stereoscopic depth (which requires that two eyes simultaneously view slightly different pictures). He recognized clear limits to the realism he could portray.

Five hundred years later the limits of depicting depth in art remain true (except of course for “Magic Eye”–style prints, which, through multiple similar elements, basically interleave two views that the brain sorts out for each eye). But Leonardo could not have anticipated the Op Art movement of the 1960s, whose chief focus was to create the illusion of movement using static images. The art form grew wildly popular in the culture at large—the mother of one of us (Rogers-Ramachandran) even wallpapered an entire bathroom in a dizzying swirl of such black-and-white patterns.

The movement never really attained the status of sophisticated “high art” in the art world. Most vision scientists, on the other hand, found the images to be intriguing. How can stationary images give rise to motion?

Psychologist Akiyoshi Kitaoka of Ritsumeikan University in Kyoto, Japan, has developed a series of images called Rotating Snakes, which are particularly effective at producing illusory motion. As you gaze at a, you soon notice circles spinning in opposite directions. Viewing the image with your peripheral vision makes the motion appear more pronounced. Staring fixedly at the image may diminish the sense of movement, but changing your eye position briefly by looking to one side refreshes the effect. In this image, you see movement in the direction that ­follows the colored segments from black to blue to white to yellow to black. Yet the colors are merely added for aesthetic appeal and have no relevance to the effect. An achromatic version (b) works equally well so long as it preserves the luminance profile of the colored version (in other words, as long as the relative reflected luminance of the different patches remains the same).

But why does this illusion arise? We do not know for sure. What we do know is that the odd arrangements of luminance-based edges must somehow “artificially” activate motion-­detecting neurons in the visual pathways. That is, the particular patterns of luminance and contrast fool the visual system into seeing motion where none exists. (Do not be alarmed if you don’t see the movement, because some people with otherwise normal vision simply do not.)

To explore motion perception, scientists often employ test patterns of very short movies (two frames in length). Imagine in frame one a dense array of randomly placed black dots on a gray background. If in frame two you displace the entire array slightly to the right, you will see the patch of dots moving (jumping) to the right, because the change activates multiple motion-detecting neurons in your brain in parallel. This phenomenon is termed apparent motion, or phi. It is the basis for “motion” pictures in which no “real” motion exists, only successive still shots.

But if in the second frame you displace the dots to the right and also reverse the contrast of all the dots so that they are now white on gray (instead of black on gray), you will see motion in the opposite direction—an illusion discovered by psychologist Stuart M. Anstis, who has been at the University of California, San Diego, since 1991. This effect is known as reversed phi, but we shall henceforth call it the Anstis-Reichardt effect, after the two vision scientists who first explored it. (Werner E. Reichardt was the founder of the Max Planck Institute for Biological Cybernetics in Tübingen, Germany.) We now know that this paradoxical reverse motion occurs because of certain peculiarities in the manner in which motion-detecting neurons, called Reichardt detectors, operate in our visual centers.

14 Comments

1. 1. mapper 11:09 AM 1/26/10

   You know - this article would have been so much better had the authors included images (or at a minumum - links to the images) they were discussing. Am I crazy, or is that a no-brainer?

   Reply | Report Abuse | Link to this

2. 2. LB in reply to mapper 11:19 AM 1/26/10

   No, you aren't crazy. A picture is worth a thousand words...except in an article that is discussing optical illusions. Then, it's worth a lot more.

   Reply | Report Abuse | Link to this

3. 3. Quasimodo 11:22 AM 1/26/10

   No brainer. What is going on over there? Some of the infinite number of monkeys have left the building.

   Reply | Report Abuse | Link to this

4. 4. heliumkidz 11:22 AM 1/26/10

   try this: http://www.itbhuglobal.org/chronicle/rotating-snakes.jpg

   Reply | Report Abuse | Link to this

5. 5. panamabob 11:24 AM 1/26/10

   Here's a link for a good demo.
   http://interactive.usc.edu/members/yuechuan/archives/optical-illusion-wheels-circles-rotating.png
   Bob
   

   Reply | Report Abuse | Link to this

6. 6. panamabob 11:29 AM 1/26/10

   Here's another:
   http://weblogs.cltv.com/entertainment/tv/metromix/optical-illusion.jpg

   Reply | Report Abuse | Link to this

7. 7. kieryn phipps 12:15 PM 1/26/10

   Actually the pictures are all there... we just can't see them because of a very clever optical illusion!

   Reply | Report Abuse | Link to this

8. 8. mapper 02:11 PM 1/26/10

   I would like to thank heliumkidz and panamabob for doing the work of the SciAm writers. Maybe they will contact you the next time they dream up an article to write!

   Reply | Report Abuse | Link to this

9. 9. dbtinc in reply to LB 02:29 PM 1/26/10

   my thoughts exactly - a picture is worth a thousand words. Author what weren't you thinking.

   Reply | Report Abuse | Link to this

10. 10. panamabob in reply to mapper 05:10 PM 1/26/10

    Thanks for the thanks, my pleasure.

    Reply | Report Abuse | Link to this

11. 11. rhodinsthinker 09:53 PM 1/26/10

    The pictures exist. The article posted on the Internet is a teaser. You have to buy the (digital) magazine to see the pictures.

    Reply | Report Abuse | Link to this

12. 12. randy@wireonfire.net 04:59 PM 1/29/10

    Does anybody have a link to the two frame movie? Thanks for the links that weren't included. SA this article cries out "multimedia" I can hear it.

    Reply | Report Abuse | Link to this

13. 13. hsabsolute 10:17 PM 1/30/10

    The reference in this article to (mechanically) detectable "neuronal activity" leaves a bit to be desired. Yes, the neurons are active during the process of sensing motion --just as the very credible observations indicate, however, simply being "active" may presume too much to be derived from the observation. In the next step, the possible malfunction accounting for the optical phenomena is noted.
    
    It seems that a preliminary step is missing in the account. This conclusion is suspiciously akin to reporting the occurrence of an earthquake, invasion or other event when it is observed as an increase in "phone calls" from the affected region.
    
    The question? What "message" (and, even more interesting, in what format?) is being sent? The neuron "signal" model will not account for sufficiently robust information flow to explain the reliability of visual sight with its incredibly diverse products.
    
    The next question: What exactly is the nature of what might be stored in memory which will ultimately provide a "match" to the visual data, rendering its identity to the conceptual mind?
    
    I'll try to pack off an interesting zip file to the authors.

    Reply | Report Abuse | Link to this

14. 14. Dom Amato 07:44 PM 2/13/10

    We would see the images if we bought the article.
    
    

    Reply | Report Abuse | Link to this