Gestalt principles describe general tendencies, not a fixed lookup-table priority

What it is

Statements like "proximity beats similarity" or "connectedness is the strongest cue" describe a statistical tendency observed under one specific set of test conditions — they are not a universal table you can look up to determine the outcome of any new layout. Treating them as a fixed, unchanging ranking is a misuse of these findings, not a flaw in the findings themselves.

Why it happens

A grouping outcome is pulled out jointly by how strong each cue actually is in a given display, so a "who beats whom" ranking is, at bottom, just the outcome that was typically observed at the particular spacing and color values used in the original tests. Change those values — push a color contrast to an extreme, or compress a spacing difference until it is barely there — and the original ranking can easily stop holding, or even reverse. So the "priority" people commonly cite describes what tends to happen when cue strengths sit at typical, moderate levels, not a law that holds by logical necessity.

Where it stops holding

This does not mean these tendencies have no value — when there is no time or budget to run a specific test and a first draft has to be produced from experience alone, these tendencies are still a more reliable starting point than guessing blind. They just should not be treated as the final answer; the draft still needs confirming against the actual layout afterward.

Applying it

• When internal design documentation or component specs cite these findings, avoid phrasing them as absolute rules like "proximity > similarity." State instead that this is a common tendency at typical cue strengths that needs confirming against the actual layout, so that someone downstream does not treat it as a lookup-table conclusion that needs no verification.
• When someone tries to use "this principle is theoretically stronger" to override a grouping result actually observed in testing, point out that the principle describes a typical tendency, while the specific values in the actual layout decide who really wins this time — an abstract ranking should not be used to dismiss a concrete, tested observation.

Related

• Same group: A2.10.1 When several principles act at once, the outcome is set jointly by each cue's relative strength · A2.10.6 Cue-conflict outcomes in a specific layout need user testing, not just reasoning from principles
• Nearby: A2.01 Proximity · A2.08 Connectedness
• Search terms: Gestalt tendency · not a fixed hierarchy · perceptual grouping caveat