Prototype
Visual Scene Change
A generated abstract scene appears, disappears, then appears again. Click the changed area, or press Same if nothing changed.
Press start when ready.
About this test
Visual Scene Change is a change-detection test. A generated abstract scene appears briefly, disappears, and then returns. Most trials contain one change. The task is to find the changed area, or choose Same when the scene has not changed.
The test is inspired by change blindness: the tendency to miss visual changes when the change is separated by a blink, blank screen, or other interruption. A change can be clear once pointed out, but still hard to catch during the actual trial.
This is not meant to be a photographic memory test. Real visual memory is not usually like storing a perfect image. The goal is to briefly hold the layout of a scene, compare it with a second view, and detect what no longer matches.
What it measures
- Visual attention
- Short-term visual memory
- Memory for object locations
- Detection of small layout changes
- Performance under visual clutter
How scoring works
The score is the highest completed level. A single run can be noisy, so the useful number is usually the range reached across several attempts.
What happens during the test
The visual system first picks up basic details: edges, contrast, shape, position, and direction. These early visual areas sit toward the back of the brain.
The scene then has to be organized into objects and layout. Some brain areas are more involved in recognizing what things look like, while others are more involved in where things are placed.
Attention matters because the whole scene cannot be stored with perfect detail. The brain has to select enough useful information to make the second scene comparable to the first.
Front control areas help keep the task goal active: compare the two views, decide whether a change happened, and choose the response. The result is a visual attention task under time and clutter.
Why it gets difficult
The change itself is designed to stay visible. Difficulty mainly comes from the number of possible details that could have changed.
Early levels have only a few objects and lines, so the scene is easy to compare. Later levels contain more anchors, connections, and small marks. The changed detail may still be clear, but finding it requires a better memory of the whole layout.
What affects results
- Screen size and viewing distance
- Brightness and contrast
- Fatigue and sleep
- Stress or distraction
- Mouse accuracy
- Scanning strategy
Further reading
Rensink, O’Regan & Clark — To See or Not to See
A classic change-blindness paper about why large changes can be missed when a blank interruption hides the motion cue.
Luck & Vogel — The Capacity of Visual Working Memory
A well-known visual working memory paper about the limits of holding visual objects in mind.
Sperling — Brief Visual Presentations
Early work on iconic memory: brief visual traces that can hold more information than can be fully reported.
Visual Working Memory in Human Cortex
Review of brain areas involved in holding visual information in mind, including visual, attention, and control regions.
Attention and Working Memory
Review of how attention and control systems support working memory, focusing, and response selection.