Cognitive Daily reports nearly every day on fascinating peer-reviewed developments in cognition from the most respected scientists in the field.
Check this out:
As you might guess, all these bands are actually horizontal, but the stripes cause each band to be perceived as sloping up or down. But take a closer look: the top pair of bands appears to be getting closer together as you move from left to right, while the bottom pair appears to be getting farther apart. Yet the stripes on each pair of bands are slanted in the same direction: the only thing that makes the bands appear to slope in different directions is the frequency of the stripes.
I found this illusion via David Whitaker's web site. Whitaker submitted one of the top ten illusions in this year's visual illusion contest. Go check them all out!
Comments
Since young, I have always found optical illusion fascinating :)
Posted by: pelf | May 17, 2007 10:28 AM
Anyone else notice that if you look at the two bars at the bottom for even a couple of seconds, then shift to the top two, you see them superimposed?
At first I thought that was part of the image, but then realized it was just an afterimage.
Posted by: Dave Munger | May 17, 2007 10:51 AM
Yep, I notice it too !
Posted by: Ben Plante | May 17, 2007 12:04 PM
What counts as "boundaries" here? I mean "boundaries that define horizontal bands." Are there boundaries in some original artwork on paper? Are boundaries defined as changes above a threshold difference from a background tone, color, etc.?
The boundaries of the bands are sufficiently fuzzy on my Dell flat screen monitor that I am unable to (e.g.) superimpose a piece of paper on the screen and convince myself that there are indeed bands that are horizontal.
Posted by: RJ | May 17, 2007 1:38 PM
It's not just the frequency of bands that differs, but the angle or slope of the bands as well. I think my brain interprets the top set as light reflected in a pair of chrome tubes...I don't know what to make of the lower pair.
Posted by: CCP | May 17, 2007 1:47 PM
I was convinced it was horizontal by the paper on screen test.
So I downloaded it, pulled it into inkscape and... it is horizontal!
I want it as part an Urban Living Camoflage / Dazzle paint clothing outfit.
Posted by: John Carter | May 17, 2007 5:25 PM
The real illusion in the top bars is that they successfully pass off as illusion. Measure the vertical distance between bars (what's visible of them) at the left edge: 53 pixels. Right edge: 43 pixels. So what you eye sees is the 10-pixel difference. It's also easy to see that the white the stripes are tilted toward each other.
Posted by: Alexei Lebedev | May 21, 2007 12:49 PM
http://illusioncontest.neuralcorrelate.com/
Forbidden
You don't have permission to access / on this server.
Posted by: Brad | May 22, 2007 10:51 AM
Sorry about that Brad.
I think they may have taken down the site -- none of the links work any more.
Posted by: Dave Munger | May 22, 2007 10:59 AM