Iris
Illusion: Müller-Lyer
Which horizontal line looks longer?
Choose Illusion

The two horizontal lines are exactly the same length. The inward-pointing arrows make the top line appear shorter, while the outward-pointing arrows make the bottom line appear longer. Your visual system interprets the arrow fins as depth cues — corners receding versus corners projecting toward you — and scales perceived length accordingly.


About this lab

Visual illusions are not failures of perception — they are windows into how the visual system processes information. Your brain does not passively receive images from the retina. It actively constructs a model of the world using assumptions built up over a lifetime of experience with physical reality. Illusions occur when these normally useful assumptions are applied to stimuli specifically designed to violate them. The Müller-Lyer illusion, first published in 1889, exploits depth cues related to perspective; the Ebbinghaus illusion exploits size contrast; the Café Wall exploits edge detection.

The Hermann Grid illusion, discovered by Ludimar Hermann in 1870, reveals how lateral inhibition works in the retina. Retinal ganglion cells have center-surround receptive fields: they respond strongly when their center is stimulated but their surround is not. At the intersections of white lines on a dark grid, the surround of a receptive field is more stimulated than along the lines themselves, causing reduced firing and the perception of dark spots. This illusion disappears when you look directly at an intersection because foveal receptive fields are too small for the effect.

The Ponzo illusion demonstrates size constancy — the visual system's tendency to perceive objects at a greater distance as larger than equally-sized objects nearby. The converging lines trigger the same depth processing that handles railroad tracks or hallways receding into the distance. The brain interprets the upper horizontal line as farther away, and since it subtends the same retinal angle as the closer one, it must be physically larger. This correction is usually accurate in the real world but produces a reliable illusion with flat images.

What makes these illusions scientifically valuable is that they are cognitively impenetrable: knowing the lines are equal does not make them look equal. This tells us the distortion occurs at a processing stage that is inaccessible to conscious reasoning. They reveal the architecture of early and mid-level vision — the computations that happen before the results reach your awareness. Each illusion isolates a different assumption, giving researchers a toolkit for mapping how the brain transforms raw retinal input into the stable, three-dimensional world you experience.