The McCollough Effect: Why These Grayscale Stripes Will Look Pink for the Next Three Months

If you stare at a bright green square for thirty seconds and then look at a blank white wall, you will see a faint pinkish square drifting across your field of view. Within half a minute, it fades. Your photoreceptors have simply replenished their pigments.

But there is a different kind of visual shift that does not fade so easily.

If you spend ten minutes alternating your gaze between horizontal green stripes and vertical red stripes, a strange change takes place. When you subsequently look at a pattern of plain, black-and-white horizontal stripes, the white gaps will appear distinctly pinkish. When you look at vertical black-and-white stripes, they will appear greenish.

Blink all you want. Splash water on your face. Walk away and look at the sky. If you return to those black-and-white stripes tomorrow morning, they will still look colored. If you limit your exposure to the test patterns, the illusion can persist for up to 85 days, as reported in IFLScience.

This is the McCollough effect, a contingent aftereffect discovered by American psychologist Celeste McCollough in 1965 (Science). It represents a fundamental recalibration of the early visual cortex, and neuroimaging studies have begun to pin down exactly where and how this structural lock occurs.

The Early Search for the Neural Locus

In standard sensory adaptation, such as the drifting square afterimage, the change is peripheral. It happens in the retina. Photoreceptors become temporarily depleted or fatigued, a process that reaches equilibrium in under forty minutes (The Illusions Index).

The McCollough effect operates differently. One early clue to its neurological origin is that it exhibits little to no interocular transfer. If you cover your left eye and induce the effect using only your right eye, looking at the black-and-white stripes with your left eye yields no illusory colors (The Illusions Index). This demonstrates that the effect must occur at or before the point where the signals from both eyes are fused.

Retinal cells do not care about orientation; a horizontal line and a vertical line stimulate them identically. Selective tuning for orientation first emerges in the primary visual cortex, or V1. This is where individual neurons fire preferentially in response to specific angles, such as horizontal or vertical edges (The Illusions Index). McCollough originally proposed that the effect was caused by the adaptation of color-sensitive edge detectors in this early cortical region (Scholarpedia).

However, early functional magnetic resonance imaging (fMRI) studies in the late 1990s muddied the waters. Researchers like T. W. James and colleagues (1999) observed stronger, more consistent blood-oxygen-level-dependent (BOLD) activation in higher-level extrastriate regions, such as the fusiform gyrus, when participants experienced the illusion (Scholarpedia). This suggested that higher-order, top-down processing might be projecting the colors backward onto the early visual regions.

Inside "McCollough World"

To resolve where the calibration actually takes place, vision scientists needed a way to safely induce an exceptionally strong version of the effect in a laboratory scanner.

At the University of Minnesota, Katherine E. M. Tregillus and Stephen A. Engel designed an augmented-reality environment they called "McCollough World." Participants wore a head-mounted display equipped with a camera. Real-time video from the camera was filtered in the Fourier domain, stripping away all diagonal visual information to pass only vertical or horizontal lines. High-contrast red was added to the vertical lines, and green was added to the horizontal lines, switching every two seconds.

By having participants navigate this modified environment for two hours, the researchers induced a robust, highly stable aftereffect (Journal of Vision).

Using high-resolution fMRI and multivariate pattern analysis, Tregillus, Engel, and colleagues mapped the brains of participants before and after their stay in McCollough World. The team trained a machine-learning classifier on the patterns of voxel activity in V1 during the pre-adaptation phase, teaching it to distinguish between physically red-green gratings and achromatic (black-and-white) gratings.

When they ran this classifier on the post-adaptation scans, the results were clear. When participants looked at physically black-and-white stripes, the V1 voxel activity patterns were classified as red or green, matching the participants' subjective, illusory color perception (Journal of Vision).

The neural signature of the illusion was directly decoded from V1. Measured.

Error Correction and the Brain's White Balance

If the mechanism sits within the earliest processing stages of the visual cortex, why does it last for months?

One prominent explanation is that the McCollough effect is not a malfunction, but a highly precise error-correcting device (Wikipedia). The physical lens of the human eye is subject to chromatic aberration; it bends different wavelengths of light at slightly different angles, creating colored fringes along high-contrast borders.

In a natural environment, structural lines are not consistently paired with specific colors. If your brain suddenly receives hours of input where vertical lines are always red and horizontal lines are always green, it does not assume the world has changed. Instead, it assumes that the eye's physical hardware has developed a pathological defect (Wikipedia).

To compensate, the visual cortex applies a semi-permanent "white balance" adjustment, subtracting green from horizontal orientations and red from vertical ones. When you look at normal black-and-white stripes, you perceive this subtraction as the complementary color.

The longevity of the effect represents a form of associative learning at the cellular level. V1 neurons that fire in response to vertical edges become bound to the color-opponent pathways that signal greenness. Because this calibration is meant to fix what the brain perceives as a physical error in the eye, it remains locked in place, waiting for counter-evidence to reset the system.

If you have induced the effect, the fastest way to clear it is not to wait three months. You must sit in front of the induction patterns again, with the colors reversed, to convince your early visual cortex that its hardware is functioning correctly after all.

---

This article is AI-generated (synthetic) content, produced by an automated editorial system with human direction and review. Every claim is traced to published, peer-reviewed sources.