Skip to content
Lizely

Change Blindness Test

Study five original abstract scenes, pass through a brief flicker, and locate the single visible change.

Privacy: your files never leave your device. All processing happens locally in your browser.

How to use

  1. 1.Study the Original abstract scene for 0.9 seconds, or press S to request the change sooner.
  2. 2.Wait through the 0.18-second blank mask, then click the changed zone or press its number from 1 through 9.
  3. 3.Find all five single-feature changes before a second distinct miss to finish with exactly 1,000 points.

About Change Blindness Test

Change Blindness Test is a five-round observation game built from original abstract SVG scenes. Each trial shows one scene, inserts a brief blank visual mask, and returns with exactly one altered feature. Locate the changed area on a three-by-three zone grid before a second distinct miss ends the run.

The scenes are fixed product fixtures rather than photographs, scientific stimuli, diagnostic material, or an imported perception test. Their rings, squares, rays, stripes, orbit dots, bridges, timing, zone map, and score rules were authored specifically for this browser game. No result claims to measure eyesight, cognition, attention health, or any clinical ability.

Every round follows the same transparent state machine. The Original frame remains visible for 900 milliseconds. A 180-millisecond neutral mask then removes the scene, after which the Changed frame appears with the nine answer zones. Show change now, or the S key, shortens only the Original stage; it still preserves the mask before the answer phase.

Round one changes the central ring into a square. Round two reduces the northwestern ray cluster. Round three rotates the southeastern stripe set. Round four moves a dot in the northeastern orbit. Round five opens a gap in the southwestern bridge. Each before-and-after pair is drawn from the same stable geometry, so only the declared feature changes.

The answer layer does not mount during Original or Flicker. Once the Changed frame is visible, each third of the scene becomes one large semantic button numbered 1 through 9 from left to right and top to bottom. Pointer and touch users can click the altered region directly. Keyboard users can press the matching number without needing precision movement.

A correct zone adds exactly 200 points and begins the next Original frame. Five correct locations therefore equal exactly 1,000 points. There is no speed multiplier, random target, timer penalty, adaptive difficulty, or hidden score adjustment. The short timers control presentation only, not how many points a correct answer earns.

One distinct wrong zone is recoverable. Choosing the same wrong zone again in the same round is recognized as a repeat and does not consume the second error. Choosing a different wrong zone records the second miss and closes the run in a terminal deadlock. Invalid zone numbers, answers before the Changed phase, and actions after completion or deadlock are harmless no-ops.

The independent logic test restates all five literal before states, after states, and correct zones rather than computing an answer from the production array. It advances the natural 900-plus-180-millisecond timer path, checks the explicit Show change transition, executes the keyboard zone route to exactly 1,000, repeats one wrong zone, triggers a second distinct miss, rejects invalid time and zone values, and verifies terminal freeze and deterministic restart.

The interface uses project surface, foreground, accent, border, danger, and muted tokens. The square scene reserves its dimensions to prevent layout shifts, the blank mask remains the same size, every answer region exceeds the minimum touch target, and visible focus and live feedback make the state understandable without relying only on color. Reduced-motion users do not depend on decorative movement.

Everything runs locally in the page. There is no upload, camera, eye tracker, account, model, third-party image, network request, or personal data. Only the shared best completed score can persist in this browser. Restart restores trial one, zero points, two available errors, and the complete Original timer.

Double-press Escape for the simulated spreadsheet boss screen, then repeat it to resume the unchanged trial. For other fixed visual challenges, try Spot the Difference, Pattern Memory Game, Hidden Object Game, or Visual Memory Test. Those games test different forms of comparison or recall while keeping their fixtures and scoring visible.

Methodology & sources

Five immutable original records declare a visual feature, before state, after state, and correct 1-based zone. A pure timer reducer advances Original for 900 ms, Flicker for 180 ms, then Answer. Correct zone adds 200 and resets the next round to Original. Wrong signatures are keyed by round and zone, so identical repeats dedupe; two distinct signatures deadlock. Invalid timing, out-of-range zones, out-of-phase actions, and terminal actions preserve state.

Frequently asked questions

Is this a medical or eyesight test?
No. It is an entertainment game using original abstract fixtures and makes no diagnostic or health claim.
Does Show change now skip the flicker?
No. It ends the Original stage early but preserves the 0.18-second neutral mask before the Changed frame.
How are repeated misses counted?
The same wrong round-and-zone choice is deduplicated; a second different wrong zone closes the run.

Related tools

Mini Games guides

View all