Skip to content
Lizely

What's Missing Game

Compare five original fixed scenes, identify exactly one removed item from explicit candidates, and complete a deterministic 1,000-point visual memory run.

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

How to use

  1. 1.Choose Start preview and study all five labeled visual items in the complete original scene for 2.4 seconds.
  2. 2.When the changed scene appears, compare the marked empty position with what you remember from the original.
  3. 3.Choose the missing item from four explicit candidates using pointer or touch, or move with arrows and press Enter or Space.

About What's Missing Game

What's Missing Game is a five-round visual comparison challenge built around one clear action: remember a complete scene, inspect its changed version, and identify the single item that disappeared. Every round begins behind a cover so the original scene never appears before you intentionally start. Starting opens a 2.4-second preview containing five visible item cards. At the exact preview boundary, the same layout returns with one position marked as gone and four explicit answer candidates below it.

The game uses five fixed scenes named Morning Desk, Picnic Cloth, Workshop Bench, Travel Shelf, and Rainy Hall. These are original Lizely fixtures rather than photographs, stock illustrations, external puzzle sets, or generated content. Each item is represented by a compact geometric mark, a two-letter visual code, and a plain English label. The combination keeps the comparison legible on small screens while making every original scene deterministic and independently testable.

Each changed scene is derived from its own original fixture by removing exactly one identified item. No second item changes position, label, shape, or code. The specific missing relationship stays inside active gameplay rather than being published in this page description. Independent tests validate every original-to-changed relationship and candidate order without placing future-round answers in rendered metadata.

Four candidates are always visible during the choice phase. One candidate is the removed original item and three are plausible but absent alternatives. You never have to tap a vague blank area, hunt through a crowded illustration, or guess where an invisible hotspot was placed. The missing position is visibly marked, and your task is to compare what was there before with the named candidate set. This makes the game distinct from Hidden Object Game and other search-style loops.

A correct comparison immediately awards exactly 200 points and prepares the next covered scene. Five correct rounds therefore produce exactly 1,000 points. There is no speed bonus, preview-skipping bonus, random multiplier, partial score, streak modifier, or hidden difficulty adjustment. Because the fixtures and score are fixed, Restart always returns to the same Morning Desk cover with zero points and zero recorded errors.

The error model is deliberately recoverable. A first distinct wrong candidate records one error but leaves the changed scene and choices visible, so you can compare again and repair the round. Choosing that identical wrong candidate again does not consume another error. The reducer stores a stable round-and-candidate signature and treats a duplicate as an atomic no-op. Only a second different wrong candidate closes the run in deadlock.

Pointer and touch players can press any candidate button directly. Keyboard players use the arrow keys to move through the two-by-two choice grid and press Enter or Space to choose. The focused candidate and a pointer click both call the same pure candidate reducer, so input method cannot change scoring, error identity, or the correct answer. Modifier shortcuts are ignored, and held activation keys are ignored after the initial keydown, preventing modified commands or native repeat from producing accidental submissions.

The candidate buttons provide at least a 56-pixel minimum height, visible keyboard focus, plain-language accessible names, and responsive two-column sizing. The scene uses two columns on narrow screens and three on wider screens, with min-width protection and truncated labels to avoid horizontal page overflow. All surfaces, borders, foreground text, accent states, and error states use the site's existing design tokens. No remote images, hardcoded color palette, canvas dependency, or downloaded asset is required.

Preview timing is protected against background-page exploits. If the document becomes hidden during the original-scene preview, that preview is canceled and the round returns safely to Ready without recording an error. Returning to the tab requires starting the complete preview again. The changed scene cannot silently appear after time expires in a hidden tab. Timing uses performance time only inside event and effect callbacks; rendering contains no Date call, random value, or changing fixture.

Only the active round is rendered. During Preview, the DOM contains the five active original items and no candidates. During Choice, it contains the active changed scene and that round's four candidates. The component never maps all five round fixtures into hidden panels, data attributes, or screen-reader-only answer lists. Future-round missing relationships therefore do not leak into the current page DOM.

Completion and deadlock are terminal states. Preview start, timer advancement, visibility cancellation, cursor movement, focus selection, and candidate choice all return the same state object once terminal. The terminal result receives programmatic focus so keyboard and screen-reader users are brought directly to the outcome. Restart is the single explicit path back to the deterministic initial fixture. The shared game shell also stores the best score locally and retains the double-Escape boss key, which swaps the game for the site's spreadsheet-style cover and restores it with a second double press.

Everything runs locally in the browser. The scenes, candidate choices, timing state, errors, score, and best-score update require no account, upload, API, paid service, device permission, or new package. The game is recreational. It does not diagnose memory loss, measure intelligence or attention, certify visual ability, or provide an educational or clinical score. A 1,000-point result means only that all five authored comparisons were answered correctly under the disclosed rules.

Methodology & sources

Author five literal scenes with five unique item IDs each, one literal missingId, and four unique candidates containing that missingId exactly once. Start each round in Ready, accept only a finite nonnegative performance timestamp, expose the full active scene during a 2400-millisecond Preview, and move to Choice only at or after the inclusive boundary. Derive the changed scene by filtering exactly missingId. Cancel Preview to Ready on document visibility loss. Route pointer, touch click, Enter, and Space through chooseMissingItem; use pure two-by-two cursor actions for arrows. Sign wrong attempts as roundIndex:candidateId, keep the first distinct error repairable, return identity for a duplicate, deadlock on the second distinct signature, award 200 for each correct round, total exactly 1000, reset round-local errors after a correct answer, freeze every action at completion or deadlock, and make Restart return the exact initial fixture.

Frequently asked questions

How do I play What's Missing Game?
Start each preview, study the five-item original scene, then choose the one removed item from the four candidates shown under the changed scene.
Is this the same as a hidden object game?
No. Nothing is concealed inside a busy picture and there are no invisible hotspots. You compare two versions of one fixed scene and select from explicit candidates.
What happens if I hide the browser tab during preview?
The active preview is canceled safely with no error. Return to the page and start that round's complete 2.4-second preview again.

Related tools

Mini Games guides

View all