The pull toward simplicity works on figure recognition and does not automatically extend to information-architecture simplification

What it is

The visual system's preference for regular, simple structures concerns how a single figure's outline and boundaries get organized into a whole — a question at the level of shape recognition. It gets invoked, often, to back a claim at a completely different level: flattening a navigation menu from three tiers to two, or merging ten function entries into five broader categories, gets called "simpler, more in line with how the brain likes things." That is a switch in levels. Simplifying information architecture deals with categorization, naming, and findable paths — problems that depend on language and logical reasoning — which is a different kind of processing from how the visual system reads a set of lines as one shape. A conclusion from the first does not transfer to justify the second.

Why it happens

Simplicity at the figure level is an automatic, fast perceptual process that needs no deliberate thought, and it runs on visual input itself. Whether something is easy to find in an information architecture depends on whether a user's own guess — "which category should the thing I'm looking for fall under?" — matches how the product actually categorizes it. That relies on language, category knowledge, and prior experience: an active, reasoning-dependent cognitive process, not passive perceptual organization. Compress three tiers into two, and if the resulting category names become broader and vaguer, users end up spending more time deciding which broad bucket their target falls into — nothing like the effortless, instantaneous judgment that happens when a figure is simply seen as one shape.

Where it stops holding

If, after collapsing tiers, the new category names stay clear and well-bounded and users' expectations still line up with the actual structure, fewer tiers really does mean fewer steps to find something — a genuine simplification. But if the collapse just stuffs previously clear categories into vaguer umbrella terms, the number of visible steps may drop while the time and error rate for deciding which category to search actually rise. "Fewer tiers, fewer clicks" cannot be taken as evidence of successful simplification in that case, because what determines findability is how clear the categories are, not the tier count itself.

Applying it

• Before flattening an information architecture, test the old and new structures separately on whether users can accurately judge which category a given function belongs to — don't just count tiers or clicks.
• When merging categories, give the merged category a specific name that covers what the merged items genuinely have in common, instead of masking the differences that still exist under a vague label like "more" or "other."
• To verify: run a card-sorting exercise — have users categorize and name a set of function items themselves without seeing the existing menu — and compare their spontaneous categorization logic against the flattened architecture. Wherever they diverge is the findability cost the flattening introduced, and it needs to be handled directly rather than assumed away because the result "looks simpler."

Related

• Same group: A2.12.1 The visual system reads ambiguous figures as the simplest possible structure — the perceptual basis of the pull toward simplicity · 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 · A2.12.6 Fewer visual elements does not mean less cognitive load — the two need to be verified separately
• Nearby: A7 Mental models · G2 Navigation structure
• Search terms: information architecture · card sorting · findability