The lilac chaser is a visual illusion in which a ring of blurry magenta or lilac-colored discs, arranged around a central fixation cross, produces three distinct perceptual effects: a green dot that appears to chase around the ring, a gradual fading of the lilac discs themselves, and eventually the impression that only a lone green spot is circling an otherwise empty grey field. Created by visual scientist Jeremy Hinton around 2005, the illusion became one of the most widely shared demonstrations on the internet because it bundles several well-studied phenomena into a single, easy-to-reproduce display. Understanding it means unpacking three separate tricks your visual system plays at once.
What You See and Why It Unfolds in Stages
The standard version of the lilac chaser places twelve blurred lilac discs evenly around a circle, with a small black cross at the center. One disc at a time disappears briefly (typically for about 100 to 200 milliseconds) in sequence, so the gap travels around the ring like a clock hand. If you simply watch the ring casually, you notice the gap moving but not much else. The interesting part begins when you lock your gaze on the central cross and hold it there.
Within seconds, the gap where a lilac disc has just vanished starts to look green rather than grey. That green phantom is a negative afterimage, the complementary color your visual system generates after being exposed to magenta. A few seconds later, a second effect kicks in: the remaining lilac discs begin to fade. They do not all disappear at once, but the ones farthest from the gap progressively wash out into the grey background. After roughly ten to twenty seconds of steady fixation, many people see nothing but a single green dot orbiting an otherwise blank field. The illusion stacks these effects on top of each other, but they arise from different mechanisms.
The Green Phantom and Negative Afterimages
When your eye is exposed to a colored stimulus for even a short time, the neural machinery handling that color becomes temporarily less responsive. Shift your gaze to a neutral surface and you briefly perceive the complementary color. In the lilac chaser, each disc in the ring adapts the neurons responding to its magenta hue. The instant that disc is removed, the adapted region rebounds and you perceive green, the chromatic opposite of magenta.
This much has been known for more than a century, but the details of where the afterimage is actually generated have been debated. A common textbook explanation credits photoreceptor fatigue in the retina. Research complicates that story. A study published in Current Biology found that for normal viewing conditions (lights well below levels that bleach photoreceptors), a post-receptor rebound response in retinal ganglion cells is the more likely source of the afterimage signal, rather than desensitization of the photoreceptors themselves.1PubMed Central. Neural Locus Of Color Afterimages Separate work on negative afterimages similarly places their formation in the post-receptoral layers of the retina.2PubMed. Negative afterimages facilitate the detection of real images
Even this retinal account is not the whole story. Research using afterimages generated under more complex conditions found that traditional accounts based purely on retinal adaptation or physiological opponency are inadequate. The authors proposed that the perceived color of an afterimage is generated after colors themselves are constructed in the visual brain, meaning cortical processing actively participates in determining what color you see.3PubMed Central. The Constancy of Colored After-Images So the green dot in the lilac chaser likely results from a chain of events that starts in retinal ganglion cells but is shaped and refined at higher levels in the brain.
Earlier electrophysiological work mapped the time course of afterimages more precisely. After a bright flash, brain responses corresponding to a positive afterimage (the same color as the original stimulus) lasted up to about two minutes. Those responses then declined and re-emerged roughly 180 degrees out of phase, coinciding with the perception of a negative afterimage in the complementary color.4Vision Research. Electrophysiological correlates of positive and negative afterimages In the lilac chaser, each disc disappears for only a fraction of a second, so you skip past the positive phase and land squarely in the negative afterimage, which is why the phantom dot looks green almost instantly.
Why the Lilac Discs Themselves Disappear
The fading of the remaining discs is a separate phenomenon called Troxler fading, named after the Swiss physician Ignaz Paul Vital Troxler, who described it in 1804. When you hold your gaze steady on a central point, stationary objects in your peripheral vision gradually vanish. This happens because your visual system is tuned to detect change. A stimulus that stays put on the same patch of retina for long enough stops generating the kind of fresh neural signal your brain treats as worth representing.
Under normal viewing conditions, your eyes are never truly still. Even when you try to fixate on a single point, your eyes make tiny involuntary movements. The most studied of these are microsaccades, small rapid flicks that happen a few times per second. Research has established a direct link between microsaccade suppression and fading. Before a peripheral target fades from awareness, microsaccade rates drop. Before the target snaps back into visibility, microsaccade rates rise again.5PubMed. Microsaccades counteract visual fading during fixation The explanation is intuitive: each microsaccade shifts the retinal image slightly, refreshing the neural response. Without those refreshes, the peripheral signal decays.
The lilac chaser is designed to exploit this. The discs are blurry, not sharp-edged, which makes them low-contrast stimuli against the grey background. That matters because Troxler fading is stronger for low-contrast, peripheral stimuli. Researchers have hypothesized that Troxler fading occurs because receptive fields in the periphery are large enough that the tiny eye movements called drifts and tremor, even smaller than microsaccades, fail to provide effective stimulation when microsaccades are suppressed.6Neuron. Microsaccades Counteract Visual Fading during Fixation A blurry lilac blob on a grey field is almost ideal for triggering this kind of fading.
Later research confirmed that microsaccades can restore the visibility of faded targets not only in the periphery but also at the fovea, the high-resolution center of vision, for targets small enough to be contained entirely within it.7PubMed Central. Microsaccades restore the visibility of minute foveal targets The fact that even foveal targets can fade reinforces the point: your visual system actively maintains perception, and when the maintenance signal drops, things disappear. In the lilac chaser, the combination of steady fixation and soft-edged stimuli stacks the deck in favor of fading.
What the Brain Does When Objects Vanish
When the lilac discs fade, the grey background does not simply go blank in a patchy way. Instead, you perceive a smooth, uniform grey field. Your brain actively fills in the missing region with whatever surrounds it, a process researchers call perceptual filling-in. This is not the brain ignoring information; it is the brain constructing a replacement.
Neuroimaging work has captured what happens in the brain during filling-in. In one study using magnetoencephalography, researchers tracked neural responses to a flickering target as it faded from perception. When the target became subjectively invisible, the measured brain signal at the target’s flicker frequency dropped significantly in the visual cortex on the opposite side of the brain. But it did not vanish entirely. A residual, frequency-specific signal persisted even though the observer reported seeing nothing.8PubMed Central. Neural correlates of perceptual filling-in of an artificial scotoma in humans This is a striking result: your brain retains a representation of the stimulus at some level even when you have no conscious awareness of it.
More recent work examined what happens across different layers of the primary visual cortex during filling-in. Researchers found that neural correlates of filling-in appeared specifically in the deep cortical layers of visual area V1, layers that are associated with feedback input from higher visual areas rather than the initial feedforward signal from the retina.9PubMed Central. Feedback to deep layers in human V1 during perceptual filling-in This suggests that filling-in is an active, top-down process. Higher brain areas essentially send a message back to early visual cortex saying “paint this region grey,” overwriting the faded stimulus with the surrounding color.
For the lilac chaser, this means the disappearance of the discs is not passive decay. Your brain is constructing the perception of a uniform grey field and projecting it back to the earliest stages of visual processing. The green phantom dot, meanwhile, keeps refreshing because the sequential removal and reappearance of discs creates transient change at the gap’s location, keeping that spot active while everything else fades.
Opponent Colors and Why the Phantom Is Green
The specific hue of the phantom dot traces back to the opponent-process organization of human color vision. In the nineteenth century, Ewald Hering proposed that color perception is built on opponent pairs: red versus green, blue versus yellow, and black versus white. This framework initially clashed with the Young-Helmholtz theory of three types of cone receptors, but neuroscience eventually reconciled the two: the three cone types feed into opponent-process channels at later stages of neural processing.10PubMed. Ewald Herings Gegenfarben. Geschichte eines Gedankens
Magenta (or lilac) is a combination of red and blue. Its opponent color, produced when the adapted channels rebound, falls in the green range. That is why the phantom is green and not some other color. If the discs were blue, the phantom would look yellow. If they were red, you would see a cyan or greenish-blue phantom. The opponent-channel architecture determines the complementary color automatically. This is also why afterimages from looking at a sunset tend to appear as a cool blue-green patch when you shift your gaze to a white wall: the warm red-orange adapts one channel, and its opponent bounces back.
How to Make the Illusion Stronger or Weaker
Several factors affect how dramatic the lilac chaser looks, and understanding them lets you manipulate the experience:
- Fixation discipline: The entire illusion depends on holding your gaze still. Glance away from the center even briefly and the lilac discs snap back into view because the eye movement refreshes the retinal image and defeats Troxler fading.
- Disc blur: Sharper, high-contrast discs resist fading longer. The standard version uses Gaussian-blurred blobs against a neutral grey background specifically because low-contrast, soft-edged stimuli fade faster.
- Eccentricity: The farther the discs sit from the fixation point (that is, the more peripheral they are), the faster they tend to fade. Moving the ring closer to the center makes the effect weaker because foveal vision is refreshed more readily by small eye movements.
- Timing of the gap: If the gap moves too quickly, there is not enough adaptation time to produce a strong afterimage. If it moves too slowly, the discs may fade before the gap reaches them, muddling the green-phantom effect. Most demonstrations settle on a rotation speed of about one position per 100 to 150 milliseconds.
- Background color: A neutral mid-grey maximizes the effect. A white or black background introduces its own contrast, which can compete with the afterimage color or slow down Troxler fading.
Display hardware matters too. On a monitor with poor color uniformity or noticeable pixel grid patterns, the subtle blurring of the discs may be compromised, making them easier for the visual system to “hold on to.” High-resolution screens with smooth color gradients tend to produce cleaner demonstrations.
Why One Illusion Involves Three Separate Mechanisms
Part of what makes the lilac chaser so popular in perception science and classroom demos is that it is not a single trick. It layers three dissociable phenomena: negative afterimages, Troxler fading, and perceptual filling-in. Each one can be demonstrated independently. Afterimages show up any time you stare at a colored patch and look away. Troxler fading happens with any static peripheral stimulus if you hold fixation long enough. Filling-in occurs routinely at your retinal blind spot, which your brain papers over with surrounding texture every waking moment.
But in the lilac chaser, the three interact in a way that feels uncanny. The sequential disappearance of discs drives afterimage formation, which produces the green phantom. Prolonged fixation drives Troxler fading, which erases the real discs. And filling-in replaces those erased discs with smooth grey, leaving the green phantom as the sole visible object. The fact that a physically empty gap becomes the most visually salient thing on the screen is a vivid demonstration that perception is not a passive recording of what is out there. It is a construction, and when the construction process is pushed in the right way, it can produce experiences that flatly contradict the physical stimulus.
Individual Differences and When the Illusion Fails
Not everyone experiences the lilac chaser the same way. Some people report robust fading within five or six seconds; others need half a minute or never see complete disappearance of the discs. Several factors contribute to this variability.
People with naturally higher microsaccade rates tend to resist Troxler fading, since each microsaccade resets the adaptation clock. Individual differences in microsaccade production are well documented, and they track with differences in how long peripheral targets remain visible during fixation.11PubMed Central. Microsaccades restore the visibility of minute foveal targets If your eyes are “fidgetier,” the discs hang around longer.
Attention also plays a role. Actively attending to the peripheral discs rather than genuinely focusing on the central cross can slow or prevent fading, because directing attention to a stimulus helps maintain its neural representation. This is why the instructions always emphasize staring at the cross and not trying to “watch” the discs. Paradoxically, the less you try to see the effect, the more reliably it appears.
Color vision anomalies change the experience in a different way. People with red-green color deficiency may see a different complementary color for the afterimage, or a weaker one, because the opponent channels respond differently. The fading component of the illusion, however, is unaffected by color vision status, since Troxler fading is driven by luminance and spatial adaptation rather than chromatic processing specifically.
Afterimage Research Beyond the Illusion
One reason afterimages attract ongoing research interest is that they open a window into how visual experience is generated. Classic experiments showed that pressure-blinding one eye during adaptation (temporarily blocking retinal blood flow to suppress photoreceptor activity) eliminated negative afterimages from weakly bleaching stimuli but left positive afterimages intact, confirming that the two types of afterimage have different neural origins.12Vision Research. Long-lasting afterimages caused by neural adaptation Positive afterimages appear to originate in the photoreceptors themselves (a lingering signal after stimulation), while negative afterimages depend on neural processing beyond the receptors.
More recent findings push the story further up the visual hierarchy. The observation that afterimage color may be determined at cortical levels, beyond simple retinal opponency, suggests that afterimages are not merely “residue” from eye fatigue but are actively constructed perceptual events.13PubMed Central. The Constancy of Colored After-Images This matters because it means the lilac chaser’s green phantom is not just your retina being tired. It is your brain building a green dot because the conditions call for one, using the same generative processes it uses for ordinary color perception. The line between “real” and “illusory” color gets blurry when the same neural machinery produces both.
The Blind Spot Connection
Every human eye has a blind spot where the optic nerve exits the retina, leaving a region with no photoreceptors at all. Under normal conditions, you never notice it because the brain fills in the gap using information from the surrounding visual field. This everyday filling-in uses the same basic mechanism that erases the lilac discs during Troxler fading. In both cases, feedback signals from higher cortical areas supply the early visual cortex with a constructed representation that replaces missing input.14PubMed Central. Feedback to deep layers in human V1 during perceptual filling-in
You can verify this yourself. Close one eye, hold a finger at arm’s length in your peripheral vision on the side of the closed eye, and slowly move it outward. At a certain point it vanishes, swallowed by your blind spot. The brain fills the gap with whatever pattern is behind the finger, just as it fills the locations of faded lilac discs with grey. The lilac chaser, in a sense, is making visible what your brain does invisibly all the time.
Using Fading Phenomena in Clinical Settings
Troxler fading and filling-in are not just curiosities. They have practical relevance in clinical vision testing. Perimetry, the standard test for detecting visual field loss in conditions like glaucoma, requires patients to fixate on a central target and report when they see flashes of light in their periphery. If a patient fixates too steadily and the test stimulus sits in one location too long, Troxler fading can cause the patient to miss a target that they should otherwise see. Modern perimeters account for this by using brief stimulus presentations and randomized locations, but clinicians still need to be aware that fading-related misses can occur, particularly in patients with unusually stable fixation.
Researchers studying scotomas (blind or dim regions in the visual field caused by retinal or neural damage) also use filling-in paradigms to understand how the brain compensates for lost input. The neuroimaging findings about persistent frequency-specific signals during filling-in are relevant here: even when a person is unaware of a stimulus because of pathological or induced scotoma, the brain retains some representation of it.15PubMed Central. Neural correlates of perceptual filling-in of an artificial scotoma in humans That residual representation raises questions about whether rehabilitation strategies could coax it back into awareness, though this remains an open area of investigation rather than a proven therapy.