How the eye decides what's the subject and what's just the stage
Designers Academy
3 July 2026 · 3 min read
In brief
The Law of Figure-Ground is a Gestalt principle describing how the visual system actively distinguishes between the main subject (figure) and its surrounding space (ground) in a scene, which designers apply to create clear, unambiguous interfaces.
Every scene splits into a figure, the thing being looked at, and a ground, the space it sits in front of. The Law of Figure-Ground is one of the original Gestalt principles, first demonstrated through reversible images where the same lines can be read as either a vase or two facing profiles depending on which region the eye assigns as the subject. The visual system cannot perceive figure and ground simultaneously as equally weighted. One has to recede for the other to come forward.
The brain does not passively receive a scene, it actively decides which regions belong to an object and which belong to the space around it, using cues like contrast, size, enclosure, and which shape appears to be in front. This decision usually resolves instantly and without effort. It only becomes a conscious process when the cues are evenly balanced, which is exactly what a reversible figure exploits. In ordinary interfaces the goal is never ambiguity. It is making sure the figure-ground decision resolves correctly and immediately, in favor of whatever the user actually needs to act on.
Figure-ground failures are some of the most common usability bugs in modern interfaces, because they are invisible to the designer who already knows what the screen means.
Check contrast first, not just for text but for the boundary between any element and whatever sits behind it. A shadow, a border, or a value difference of even a few percent is often enough to fix an ambiguous edge. Be especially careful with overlays and dark themes, where designers testing in bright, high-contrast conditions can miss a figure-ground problem that only shows up on a dimmed screen or in direct sunlight.
An interface with no clear figure has no subject. Everything competes, and nothing wins.
When figure and ground are ambiguous, users cannot tell what is clickable, what is decorative, and what they are supposed to be looking at right now. That uncertainty does not read as confusion in a user's mind. It reads as the product being broken, because the failure sits below the level of conscious analysis where users normally give an interface the benefit of the doubt.
Key takeaways
