What: Assuming that you’ve sat with this for a bit before going straight to the reading, there’s about a 50/50 chance that the “effect” happened, which is essentially your brain switching the rotation direction (even though on our side it’s the same way the entire time).

There’s a similar experiment on the site, the Spinning Duck, which is based on the silhouette illusion [1], the same trick as the famous spinning dancer.

How To: This, again, can be put down to this type of visual, that doesn’t give any explicit information regarding what is in front and behind (technically there is a tiny bit with colors, but with the dots it’s tricky for your brain to clock them).

I’m not sure the easiest way, but generally when most of the dots are in center (things like tails back, etc) you can kinda loosely focus on it going the other way and maybe get it.

Sometimes for me, looking at something else in the room with the spinning object in the peripheral I will see it instantly switch.

If you click, the model will change, so you can explore others… some I find easier, but I again can not seem to pick a reasoning behind it. The Haunter model works well for me, and I can mentally seemingly change its direction easily.

Explain it: Every dot is dropped straight onto the screen with no perspective, so a dot on the near side moving right looks exactly like a dot on the far side moving left. Your brain is very good at rebuilding a 3D shape out of moving dots, an ability called the kinetic depth effect [2], but it has to guess which way the depth goes, and both guesses fit the picture perfectly. So it picks one, and every now and then it picks the other. Same family as the Spinning Duck and Schröder’s Stairs, where a staircase does the same flip without moving at all.

Fun fact: the spinning dancer got passed around as a “left brain vs right brain” test. It’s nothing of the sort, and the fact that you can flip it on purpose rather gives the game away.

Cites and Extras:

• [1] The Spinning Duck on this site is the silhouette version of the same illusion, and the Spinning Dancer by Nobuyuki Kayahara is the famous one.
• [2] Wallach, H. & O’Connell, D. N. (1953). The kinetic depth effect. Journal of Experimental Psychology, 45(4). The original paper on building shape from motion.
• Models, all free: Pikachu by Tipatat Chennavasin (CC BY 3.0); Crewmate by J-Toastie (CC BY 3.0); Pickle Rick by Polygonal Mind (CC0); R2-D2 by Eric Finn (CC BY 3.0); Bulbasaur by Tipatat Chennavasin (CC BY 3.0); Duck by Sony (SCEA Shared Source, © 2006); Knuckles by Robert Ray (CC BY 3.0); Charmander by Tipatat Chennavasin (CC BY 3.0); Astronaut by Poly by Google (CC BY 3.0); Squirtle by Tipatat Chennavasin (CC BY 3.0); Eevee by Tipatat Chennavasin (CC BY 3.0); Snorlax by Tipatat Chennavasin (CC BY 3.0); Mew by Tipatat Chennavasin (CC BY 3.0); Bunny by Stanford Computer Graphics Laboratory (Stanford scan); Haunter by Tipatat Chennavasin (CC BY 3.0); Cubone by Tipatat Chennavasin (CC BY 3.0).

I've researched these optical illusions in my spare time but am clearly not any kind of expert and my explainations are pretty smooth brained, if you find something mis-cited, earlier examples, or general mistakes please new let me know via toymaker@toms.toys, be kind!