Nothing in This Picture Is Moving

A spiral, twenty seconds, and a photograph that starts to breathe. The photograph never changes.

This article contains affiliate links. If you buy through them, we may earn a small commission at no extra cost to you.

You watched a spiral turn for about twenty seconds. Then a still photograph came back, and it swelled, or shrank, or crawled across itself. It did none of those things. Not one pixel changed while you were looking at it.

That gap between what the screen showed and what you saw has a name. It is called the motion aftereffect, and it is one of the oldest recorded tricks your visual system plays on itself, and one you can trigger any time with nothing more than a spinning spiral desk toy and a blank wall.

Your brain does not have one detector that reports motion. It has populations of cells that each prefer a particular direction. Some fire hardest for things moving up, others for things moving down, others for clockwise, others for counterclockwise. What you experience as "movement" is the balance between them, the way a tug of war is decided by which side pulls harder.

Stare at a spiral turning one way for long enough and the cells tuned to that direction start responding less. They are not damaged and you are not tired. They are doing what nerve cells do when a signal never changes: they turn down the volume. Researchers using brain scans have found this direction specific fading in the human motion area MT, and also in earlier visual areas V1 and V2 (Huk, Ress and Heeger, *Neuron*, 200100452-4)).

Then the spiral stops. The cells you exhausted are still quiet. The ones tuned to the opposite direction are not. For a few seconds, one side of the tug of war is pulling alone, and a photograph that is sitting perfectly still gets reported to you as moving.

In 1834 a travelling lecturer named Robert Addams stood at the Falls of Foyers near Loch Ness and watched the water come down. When he turned his eyes to the rocks beside the waterfall, the rocks appeared to climb upward. He wrote it up for a scientific journal, and the phenomenon eventually picked up the nickname the waterfall illusion (*Perception*, 1994).

Addams also offered an explanation, and it was wrong. He thought his eyes had been drifting downward without his knowledge, tracking the falling water, and that the rocks only looked like they were rising because his gaze kept snapping back. It is a reasonable guess. It is also not what is happening, which is why you can produce the same effect while keeping your eyes locked on a fixed point, as you just did.

He was not even first. The description is usually traced back to Aristotle, around 350 BCE, who wrote about looking away from a fast flowing river and seeing the bank slide the other way. If that attribution is right, humans have been noticing this for roughly twenty four centuries.

Here is the part that makes this more than a party trick.

There is no cell in your head whose job is to report "this is not moving." Stillness is not a message your brain receives. Stillness is a tie, a balance between opposing populations that happen to be pulling equally hard. You do not perceive the absence of motion. You perceive a draw.

Which means that when you tilt the balance, as the spiral did, stillness stops being available as an answer. Your visual system is not malfunctioning during the illusion. It is reporting exactly what its own arithmetic says, and its arithmetic never included a category for "nothing." The illusion is not a glitch in the machine. It is the machine, briefly visible.

This is also why the effect is so hard to argue with while it is happening. You can know the photograph is frozen. You can be the person who made the video. It still moves.

Some people get a dramatic version of this. Others get a faint shimmer. A few report nothing at all.

That variation is not new and it is not a failure on your part. Large individual differences in how long the aftereffect lasts have been reported since at least the 1920s, including one documented observer who apparently never experienced any motion aftereffect whatsoever (Morgan, bioRxiv, 2018). Whether that reflects a real biological difference between people is still an open question.

Two things reliably make it stronger. Longer adaptation helps, though with diminishing returns: the length of the aftereffect appears to grow roughly with the square root of how long you stared, so doubling your staring time does not double the payoff. And keeping your eyes genuinely fixed on one point matters more than most people expect, because letting your gaze wander spreads the adaptation across different patches of your visual field instead of concentrating it.

Motion is one channel among many, and every one of them can be pushed off balance the same way. Colour, brightness, size, depth: each is computed by populations that compare rather than measure, and each has its own aftereffect waiting behind it.

Depth is the strangest of the set, because your two eyes see slightly different images and your brain silently reconciles them into one three dimensional world. The 1990s autostereogram craze was built entirely on hijacking that process, and the original Magic Eye collection still works on the same principle: uncouple your focus from your convergence, and an object rises out of a page of noise.

None of it is a defect. All of it is the price of a visual system fast enough to be useful, built by comparison rather than measurement.

The spiral just made the price visible for a few seconds.