The visual system reads ambiguous figures as the simplest possible structure — the perceptual basis of the pull toward simplicity

What it is

When a figure is genuinely ambiguous — it can be organized several different ways — the visual system does not waver evenly between the possibilities. It settles reliably on whichever overall structure is simplest, most regular, and easiest to describe. This tendency is the law of Prägnanz, sometimes called the "good figure" tendency. Interface minimalism often invokes it as a justification, but the law itself only describes how the visual system resolves ambiguity — it makes no claim that simpler looks better, or that fewer elements make a product more usable. Those are separate questions the law does not answer.

Why it happens

A contour, a gap, an overlap between shapes can in principle be organized many different ways: which part reads as foreground, which edges belong to the same object, whether a gap should be filled in as a single continuous shape. The organization the visual system settles on is typically the one that needs the fewest extra assumptions and can be described with the fewest rules — regularity, symmetry, and continuity all make a description cheaper, while skew, missing corners, and local asymmetry usually need extra clauses to spell out. The visual system favors the cheaper description because it reaches a stable, usable result at lower processing cost — not because regular shapes are inherently more attractive.

Studying it

The classic paradigm presents genuinely ambiguous figures — a regular shape partly occluded so it can be read either as one whole shape hidden behind an obstruction, or as several irregular fragments placed side by side — and asks participants to report which organization they actually perceive. Symmetry, gap size, and contour continuity are manipulated as independent variables; the proportion of participants reporting one organization over the other, and their response time, serve as dependent measures. Later work pushed further, turning the intuitive notion of "simplicity" into a computable description length: how many rules, how many independent parameters does it take to specify a structure, with fewer parameters counting as simpler. That quantification turns the law of Prägnanz from an intuitive description into a model that predicts how a specific figure will be read, rather than leaving "which reading is simpler" to a researcher's judgment call.

Where it stops holding

The law addresses how a single figure or a local set of contours gets organized into one whole; it operates at the level of shape and figure recognition, and does not by itself predict how several separate elements across a whole layout should be grouped — that depends on the relative strength of cues like spacing, color, and connecting lines. When two or more "simple" readings score close to each other — a figure that is simple along one symmetry axis and equally simple along a continuous contour — the outcome becomes unstable, and the same person may report different readings at different times. The visual system cannot be assumed to converge on one fixed answer in that situation.

Applying it

• Where users need to recognize an icon or graphic quickly and consistently as the same object, keep its outline regular, symmetric, and continuous, minimizing parts that could be organized into more than one shape — this matters especially at small sizes, where detail gets lost first.
• Do not use this perceptual tendency as a blanket justification for stripping down an entire interface — it only answers what shape a single figure will be seen as, not how many modules a page should carry, which is an unrelated question.
• To verify: show the isolated graphic with no surrounding text or context and ask several people to describe what shape they see independently. High agreement means the current outline's simplest reading is clear enough; scattered descriptions mean the outline is genuinely ambiguous and needs adjustment.

Related

• Same group: A2.12.2 Visual minimalism as a style is not the same as perceptual-processing simplicity · A2.12.3 Stripping visual cues for a cleaner look can raise the cost of discovering an interaction · A2.12.4 Forcibly simplifying the appearance of inherently complex content shifts that complexity onto the user's own mental processing · A2.12.5 The pull toward simplicity works on figure recognition and does not automatically extend to information-architecture simplification · A2.12.6 Fewer visual elements does not mean less cognitive load — the two need to be verified separately
• Nearby: A2.11 Symmetry · A2.10 Conflict and priority among Gestalt principles
• Search terms: law of Prägnanz · good figure · perceptual simplicity