What a mid-air move means is evidence from the target users, not a designer's hunch

What it is

What a mid-air movement “means” cannot be settled by a designer finding it natural. Evidence is what target users produce before they have seen the official answer, and whether they retain it after. Designers routinely stamp a function onto a movie wave, onto their own habit, or onto a shape that is easy to recognize. All three can fight the target group's metaphor, and can mean something else in the target culture. Validation is not asking a colleague “does this look like delete.” It is elicitation or guessing in the target setting, then looking at agreement.

Why it happens

Pairing gesture and function is a cultural convention, not a law of physics. The same downward wave is “off” to one person, “next page” to another, “put it down” to a third. The designer's sample size is one, further polluted by implementation constraints: easy-to-recognize shapes get the functions the team most wants. Elicitation lets many people pair functions with moves; high agreement means the metaphor already lives in the group. Low agreement means the meaning needs teaching and is not “natural.” If target users are not the designer's colleagues—age, job, culture—colleague votes are void. Validation also includes refusal: moves the group will not perform, however recognizable, cannot be kept because the designer likes them.

Studying it

The standard is gesture elicitation: give a function, hide the existing UI, let people move, compute pairwise agreement. Then guessing: give a move, let people name the function. Both must be done with the target group in the target room, not with laboratory colleagues. Log conflicting cultural meanings and reasons for refusal. High elicitation agreement still needs a separate recognition test; that is another dimension. This one only asks whether the meaning is this group's. A designer proposal can sit as a control, ranked against elicited results.

Where it stops holding

A brand-new function has no cultural metaphor, elicitation will scatter, and a taught item is legitimate—if you admit it is not natural and offer an always-available relearning door. Expert conventions may differ from the public; validate per claimed population, and do not let expert agreement vouch for a public vocabulary. Legally or safety-mandated shapes (a particular emergency-stop form) can skip elicitation, but still must be tested for whether the target group can and will produce them. Taking elicitation results as the vocabulary ignores recognition neighbours; size and neighbour columns still have to close.

Applying it

• Vocabulary reviews must attach elicitation or guessing data from the target group. A designer saying “it feels natural” is not evidence.
• Low-agreement items either become taught, with relearning, or are replaced by a higher-agreement move. Do not use copy to call low agreement natural.
• Change market or age band, remeasure meaning. Do not take one region's wave semantics as a global default.

Related

• Same group: C4.14.1 The gesture set must be bounded and discoverable · C4.14.2 Similar shapes must not carry very different consequences · C4.14.4 Learned gestures must be relearnable on demand
• Adjacent: C4.05 Three dimensions of gesture evaluation · C3.34 Cultural and regional gesture differences
• Search: gesture elicitation · agreement rate · guessability