Ebbinghaus illusion still raises strange questions.

Draw two circles exactly the same size. Surround one with big circles and the other with small ones. The first now looks smaller. Nothing about the two central circles has changed — only their company has. This is the Ebbinghaus illusion, and it has been quietly embarrassing our confidence in our own eyesight for well over a century.

The effect is named for Hermann Ebbinghaus, the German psychologist (1850–1909) best known for his pioneering studies of memory, who first described it. It reached the English-speaking world through Edward B. Titchener’s 1901 textbook of experimental psychology, which is why it is also called Titchener circles. The recipe is simple: identical centers, different surrounds, and the circle surrounded by giants shrinks before your eyes.

Why? The leading account is size contrast: the visual system judges size relatively, not absolutely, the way a person looks shorter standing next to a basketball player. But the story is richer than that. Researchers have shown that two other factors matter enormously: how close the surrounding circles sit to the center, and how complete the ring is. Bring the inducers close and the center looks bigger; move them away and it looks smaller — an effect that links the Ebbinghaus illusion to its cousin, the Delboeuf illusion. The surround’s size and its distance are tangled together, which kept researchers arguing for decades.

The illusion earned a starring role in one of neuroscience’s great debates: do the brain’s pathways for seeing and for doing work separately? The two-streams hypothesis proposes a ventral stream for conscious perception and a dorsal stream for guiding action. Neuroscientist Melvyn Goodale and colleagues had people reach out and grasp the central disc of a physical Ebbinghaus display — and the opening of their fingers was scaled to the disc’s true size, unaffected by the illusion. The eyes are fooled; the hand is not. Other studies have pushed back, finding conditions where action is fooled too, and the argument continues — which is exactly what a good illusion should do.

Individual differences run deep. Neuroimaging work has found an inverse correlation between how strongly someone falls for the Ebbinghaus illusion (and the related Ponzo illusion) and the surface area of their primary visual cortex — more cortical real estate, less fooling. Children under ten, less sensitive to context, are less deceived than university students. One recent study even linked 70 genetic variants to susceptibility. Your vulnerability to two circles is, measurably, part of who you are.

The illusion even has a rebel. In 2013, Italian visual researcher Gianni A. Sarcone discovered a relative-size illusion that runs the Ebbinghaus effect in reverse: his central test shape — a cross rather than a circle — surrounded by large squares appears larger instead of smaller. The exception does not break the rule so much as sharpen it. Change the shape, the spacing, the inducers, and the brain’s size computation shifts. The machinery is not a single trick but a whole committee of them, voting.

In 2014, researchers at the University of Nevada, Reno won the Best Illusion of the Year Contest with an animated Ebbinghaus — the illusion set in motion. More than a century after Ebbinghaus described it, the two circles are still teaching the same lesson: seeing is comparing, and every comparison can be rigged.