LLOS.ai
Gestalt & Perception
● Gestalt & Perception · Module 4 · Grouping

Figure & Ground

Before you can read anything, your eye makes one silent decision: what’s the object — and what’s just behind it.

Every scene splits in two. One part leaps forward as the figure — the thing, the subject, the shape you attend to. Everything else drops back as the ground — the backdrop it sits on. This split happens before recognition, before thought; it is the very first move perception makes. And it isn’t always settled: sometimes the same picture flips, so a vase becomes two faces, then a vase again, and you can never quite hold both at once.

For a designer this is the deepest lever there is, because whatever wins the figure wins the attention. Make the right thing the figure — a button, a headline, a dialog — and the eye goes straight to it; let everything compete and nothing leads. The tool that decides it is often the quietest one: the ground itself, the negative space you never realised you were designing. You’ll feel all of it in three quick labs.

~7 min3 felt labsa figure that flips10-question bankglossary · FAQ
The journey — Gestalt & perception
proximitynear = together
similarityalike = together
closuremind fills gaps
figurewhat is on top
continuitythe eye flows
common fatemove together
One idea runs through all six: your mind groups the world before you decide to look — and design can steer it.
What you’ll be able to do
01 · The eye picks a side

A vase — or two faces?

Here is the most famous ambiguous image in all of perception: Edgar Rubin’s vase. Look at the middle and you see a vase. Look at the edges and two faces appear, staring at each other. The picture never changes — what changes is which part you cast as the figure, and which you let fall away as ground.

Try to see both at once. You can’t. Your visual system insists on choosing one figure at a time, and it will flip — vase, faces, vase — but never hold them together. That stubborn either/or is figure & ground laid bare.

The Reversible Figure
One picture, two readings. You can hold only one at a time.
▶ Flip between the vase and the faces. The shape is identical both times — only your choice of figure changes.
The vase. A dark vase sits in the middle, figure against a light ground. Now switch to the faces — the very same edges become two profiles, and the middle drops back to nothing.
02 · No contrast, no figure

The figure you can’t see

We imagine a figure simply is a figure. But it only exists by contrast with its ground — take that contrast away and the figure vanishes, even though it’s right in front of you. Look at the dark icon below. It seems plain, with maybe the faintest hint of something.

That faint something is a right-pointing arrow, drawn in almost the exact tone of the square around it. Your eye can’t lift it off its ground, so it stays all but invisible. Give it one jolt of contrast and it leaps forward as an unmistakable figure — and you’ll never lose it again. That jolt is the whole difference between figure and ground.

The Hidden Figure
A figure is hiding in this icon, almost the same tone as its ground. Reveal it.
▶ Study the dark icon — can you just make out the shape buried in it? Then reveal it, and watch pure contrast turn ground into figure.
Looks like a plain dark icon? There’s a right-pointing arrow in there — but it’s nearly the same tone as its ground, so your eye can’t separate the two. A figure with no contrast against its ground simply isn’t seen.
03 · Design the ground

Recede the rest, and one thing pops

Here’s the payoff. On this little screen, everything competes: a stack of elements, all roughly equal, and one button lost among them. Nothing is clearly the figure, so your eye wanders.

Now push the ground back — dim everything except the one thing that matters. Suddenly the button lifts off the surface and becomes the undisputed figure; your eye snaps to it before you’ve decided to look. You didn’t change the button. You changed the ground.

Pull Focus
Same screen, same button. Watch the ground decide where you look.
▶ Toggle Focus. When the ground recedes, one element becomes the figure — and grabs your eye.
Flat. The button is here, but it’s just one more element on an even field — no clear figure, so your attention drifts. Switch to Focus and watch the ground do the work.
The craft of the ground

Whoever wins the figure wins the eye

Every lab here is one law wearing different clothes. Name it, and the background stops being scenery and becomes a control.

Decide what is figure. Push the rest to ground.

Figure and ground is the first cut perception makes, before it recognises a single thing — and it’s decided by cues you can control. A region tends to become the figure when it has more contrast, when it’s smaller or enclosed, when it’s convex, when it sits lower, and when it looks meaningful. Stack those cues on the element that matters and the eye finds it instantly; spread them evenly and you get a flat field where nothing leads. Attention isn’t won by shouting louder — it’s won by being the clear figure against a quiet ground.

Which means the ground is a tool, not a leftover. The most powerful focus techniques barely touch the figure at all — they act on everything around it. A modal dims the page behind it; a spotlight darkens all but one control; a scrim drops behind text so words lift off a busy photo; whitespace isolates a headline so it can breathe. In every case you make one thing the figure by pushing the rest back. And the ground can carry meaning of its own: the arrow in the logo, the profile in the vase, the shape you only notice when someone points to the empty space.

The discipline is restraint, because a figure only exists against a ground. Try to make everything a figure — every element bold, boxed, coloured, elevated — and you destroy the very contrast that focus depends on; the screen becomes noise. So the test is simple: on any layout, ask what the single figure is. If you can’t answer in a glance, neither can your user — and something needs to recede.

Cast your vote

Which one groups better?

Five face-offs. Same content, two spacings — one makes the relationship obvious, one makes you guess. Pick the clearer, then see why.

Round 1 of 5Score 0
Seen in the wild

Figure over ground, everywhere

Once you can see the split, you catch designers pushing things back and pulling things forward on every screen. Five ways the ground gets put to work.

Where the figure leads next

Once the eye has a figure, it wants to follow it. Give it a line, a curve, an alignment, and attention will travel — flowing along the path rather than stopping at a single shape. How the eye follows a direction, and reads separate marks as one continuous line, is continuity, and it’s Module 5.

Take it further

The real references

Figure and ground has been studied for over a century, since Edgar Rubin’s vase. Five worth your time. (Opens in a new tab.)

The vocabulary

The Figure & Ground glossary

Six words that turn “it just stands out” into a precise, usable idea. Tap a card to flip it.

The nuance

Questions you actually ask

Busting the myths

What people get wrong

Myth
“The background is just decoration.”
The ground is a control. Dimming it, blurring it or clearing it is often the strongest way to make one element pop — the focus techniques that work best barely touch the figure at all.
Myth
“More contrast and more elements grab more attention.”
Only if there’s a clear split. Make everything bold and elevated and you destroy the contrast focus needs — if everything is a figure, nothing is, and the screen reads as noise.
Myth
“Negative space is wasted space.”
The ground has a shape too. It defines the figure, gives it room to breathe, and can even be a figure itself — the arrow in the logo, the profile in the vase.
Myth
“The figure is just whatever’s biggest or brightest.”
Not always. Smaller, enclosed, convex or meaningful regions can win the figure over larger ones — and in a genuinely ambiguous image, the reading simply flips.
Train the eye

Read the grouping

Ten questions in seven formats, tagged by thinking level — each with layered hints: a nudge, the reasoning, then a deeper connection. A wrong answer opens a door, never a dead end. Served from the question bank, never hardcoded.

Loading the question bank…
Question 1 of 10
Multiple choice
Keep going

Where this connects

▶ Play the labs