A mid-air press has no bottom-out signal and needs another channel

What it is

Poke a floating OK and the fingertip goes through while the person is still waiting for the click of bottoming out. A physical key ends its travel against a hard stop. That kinesthetic end-stop timestamps the commit. Air has no stop, so a press has no completion event. A mid-air press only works if another channel — pinch, trigger, a tap sound, a visible depression, a dwell — stands in for the missing collision with a shell.

Why it happens

A real button gives travel, rising resistance, and a force discontinuity. The hard stop is a proprioceptive event: muscle meets an immovable, the instant is sharp, and the person knows the press counted. A mid-air press is isometric or unconstrained displacement; the fingertip keeps going and the force curve never kinks. People invent the missing event in two ways: push deeper and treat the clip-through as a wall, or hover until the picture changes and treat vision as the stop. The first overshoots into whatever sits behind; the second turns a click into a wait.

The substitute has to be a timestampable discrete event. A pinch moves the stop into the finger pads — real skin contact, just not with the button. A trigger has mechanical travel. A short click and a depressed caption are cross-modal timestamps, coarser than bone hitting a shell, but an event. Dwell is the weakest: it turns “press” into “wait,” and the person cannot tell a commit from a counter still running.

Studying it

A virtual-button task on a floating panel, comparing depth-of-penetration alone, added click sound, added depression animation, and pinch or trigger as the commit. Overshoot depth and whether people can say “I just pressed” tell you more than raw click rate.

Independent variables: hard stop present (physical key vs mid-air), substitute channel, button thickness, whether penetration is allowed. Dependent variables: overshoot depth, press duration, repeated taps, certainty of commit, false rejects (pressed again because it “didn't take”).

Ask immediately after each press whether it succeeded, then align with the log. “I thought it took” / system silent is the signature of a missing end-stop.

Where it stops holding

A controller trigger already supplies the stop, so a floating widget can be thin and does not need a fake one. Align the widget with a real desk or wall and the physical surface returns the stop — the premise of the leaf is gone. Dwell still helps people who cannot make a displacement, but that is a different action, not an end-stop substitute. On optical see-through, when a widget sits on a real table, people press the table; if the system is still waiting for penetration depth, the two clocks disagree.

Applying it

• Do not commit a mid-air button on fingertip depth alone. Pair it with a pinch, a trigger, or a short audio-visual event.
• Give the button a pressed visual distinct from hover, on the commit frame, so people are not guessing from how far they went through.
• When the click must be precise, snap the panel to a touchable surface or switch to pinch confirm, rather than thickening a virtual key to fake travel.
• How to check: ask “how did you know that press counted?” If the answer is “I went through it” or “I waited,” the substitute channel is not in place. A working answer points at a click, a depression, or a pinch.

Related

• Same group: N2.09.1 The contact volume must be larger than the object's visual hull · N2.09.2 The moment of release is harder to judge than the moment of grasp · N2.09.4 Vision and sound can fake a touch that is not there
• Nearby: N2.04 Direct Grab · N2.02 Bare-hand Tracking
• Search terms: kinesthetic end-stop · mid-air button · bottom-out