Continuity and closure together resolve toward whichever reading is simplest overall

What it is

When two shapes overlap or partially occlude each other, two readings often both make sense at once: one continues a contour along its original direction, passing behind the other shape (the continuity reading); the other treats each shape as independent and complete, with the hidden part filled in to make a closed whole (the closure reading). When both are available, what people actually see is whichever reading is simpler to describe — not because continuity inherently beats closure or the reverse, but because whichever reading needs fewer assumptions filled in and produces more regular shapes wins.

Why it happens

"Simpler" here is not a vague adjective — it can be counted. Take each candidate reading and tally how many separate shapes it requires, how many corners or turns are in it, and whether its edges are symmetric or regular: fewer shapes, fewer turns, and more regularity all make a description "shorter." The visual system favors whichever version can be described most economically — a preference over coding cost, not a contest between continuity and closure as competing forces. The two principles are really two candidate routes to the same goal: under some occlusion arrangements, continuing a contour smoothly happens to be the shorter description; under a different arrangement, treating each shape as its own closed whole is shorter instead. Whichever route yields the shorter description is the one that wins.

Studying it

The way to test this is to construct a set of ambiguous overlap/occlusion figures and ask participants to describe what they see, or to pick among several candidate readings, while separately scoring each candidate's objective complexity — counting how many shapes it needs, how many turning points, whether it has symmetric structure. The question is whether participants' actual choice matches the reading with the lowest complexity score. Quantifying coding cost this way and using it to predict perceptual outcomes has an established methodological line within structural-information-theory approaches to perceptual organization; the core manipulation is to widen the complexity-score gap between candidate readings step by step and check whether participants' choices track the score.

Where it stops holding

This way of quantifying "simpler" applies mainly to figures with clear contours whose ambiguity comes from overlap or occlusion — line drawings, overlapping shapes in an icon. It does not transfer directly to ordinary interface elements like rectangular cards or text blocks: that kind of layout has no ready-made objective metric for counting turning points or symmetry, and forcing the same quantitative account onto it produces conclusions with no real basis.

Applying it

• When designing an icon or illustration that implies overlap or occlusion, sketch out both candidate readings first and count how many separate shapes each needs, how many turns it has, and whether it is regular or symmetric. Choose the overlap arrangement where the reading you actually want is also the one with fewer shapes and more regularity, rather than designing against this tendency.
• To verify: strip away color and labels, leaving only the outline, and show it to someone unfamiliar with the design, asking them to describe what they see. If they describe the unintended reading, adjust how much the shapes overlap, or add a small local visual break — a gap, a shading change — to tip the balance away from the tied reading.

Related

• Same group: A2.10.1 When several principles act at once, the outcome is set jointly by each cue's relative strength · A2.10.5 Gestalt principles describe general tendencies, not a fixed lookup-table priority
• Nearby: A2.03 Continuity · A2.04 Closure
• Search terms: minimum principle · Prägnanz · simplicity principle