In everyday life the brain takes two separate images, one from each eye, and merges them into one mental image. Furthermore the brain interprets the differences in what is seen by the left eye and the right eye to gauge the distance to objects. It is automatic and pretty accurate. Stereo-pairs are two separate images taken from slightly different angles that can be viewed with a special device, that convince the brain that it is looking at a 3D scene with several different depths. Or if you are good at the magic eye things you can see a 3D image with the below stereo-pair (taken from Wikipedia) The Magic Eye images work similarly, except rather than two separate images, they start with a simple repeating pattern, and then shift a portion of the pattern horizontally relative to the pattern around it. 1_________2_________3_________4_________5_________6 When the right eye is looking at column 4 and the left eye is looking at column 3 (which can be done by aiming your eyes at an imaginary point behind your computer screen) the differences is the images received by your eyes is automatically converted to a sense of depth. As in this case that appears as a raised heart floating in front of the screen. If instead you aim your eyes at a point in front of the screen (AKA cross your eyes) such that your right eye is looking at column 3 and the left eye is looking at column , the relative differences in the images will be interpreted as a floating square, with a heart-shaped depression in the middle of it. Which is what snookums is seeing. (If you aim the right eye at column 2 and the left eye at column 4 you'll see a different image, like a floating 3 lobed duck foot here or a triple-peaked volcano in the other days image) Usually when the eyes point at something they focus of that thing as well, and vice versa. Here you need to aim you eyes at a point behind the screen (or in front of the screen) while having your eyes focus at the depth of the screen. It is a learn-able behavior, but it is somewhat subverting the usual automatic operation of the brains vision mechanism. It is akin to trying to bend your ring finger while keeping your pinkie straight (or vice versa) or trying to pat your head with one hand, while you rub your belly with the other. The actions are done together so frequently that the brain takes a short cut, and usually sends both signals together.

I had more trouble with these trying to make them pop and be clear images. Did not work so well.

When it goes into 3 d they aren't eyes they are just the pointy center spike that's in the others. First row multiple rings with center spike. Second row less rings with center spike. Third row ring and spike. Like more forceful drop in the pond to least forceful.

That's a pretty one. At first I thought it looked like a sweater on a woman's chest. Hahaha