A device-fixed anchor requires the user to remember and return to the same physical location

What it is

If the interaction volume is device-fixed, valid input always lives in the same physical patch in front of that screen or sensor. After walking away, sidestepping, or sitting in another chair, the user must return to that place to count again. The system will not move the box because the person moved. What is remembered is room coordinates, not “the air in front of my chest.”

Why it happens

The containment test uses screen or room coordinates; a person’s translation does not rewrite the box. That is natural at a ticket kiosk or a museum case: the device cannot walk, so the operator should stand at the device. The cost is spatial memory. With no mat on the floor, people estimate whether they are still inside from how the screen looks—stepping back to see the whole image often exits the box. The next empty reach is then spent on changing the gesture, not the stance. Floor marks, carpet edges, and queue lines can externalize the memory; with none of those, a device anchor turns “go back to the spot” into hidden work.

Studying it

Have participants finish a task inside the volume, then send them away to look, talk, or make room, and ask them to continue without a stance hint. Record return path, distance from the stopping point to the volume, and whether they ask “do I have to stand back there.” Compare floor marks, an on-screen stance cue, and no cue. The measurement is spatial memory and prompting, not the gesture vocabulary itself.

Where it stops holding

In a cockpit, dental chair, or checkout whose seat cannot move, people were never going to leave the device anchor, so this memory load is near zero. A portable tablet with a front camera moves the device: “fixed to the device” now travels with the tablet, and what must be remembered is position relative to the slab, not a floor tile. Strong lighting or a floor pattern that already suggests a stance makes the no-mark condition behave like a mark. When an accessible route changes where a body can stand, the device anchor may sit in space that cannot be occupied—remembering does not get you back.

Applying it

• Give a device-anchored product a recoverable stance on the floor or in the UI; do not assume people will feel their way back to a centimetre-scale original spot.
• When a reach fails after the person has left, point the hint at stance (“please stand in front of the screen”), not at gesture shape first.
• Walk “finish, leave, come back and continue”: count how many first reaches land outside the volume. That count is the anchor’s memory cost.

Related

• Same group: C4.21.1 The coordinate anchor decides whether the volume is device-fixed or relocates with the body · C4.21.3 A body-anchored volume keeps the operating experience consistent after the user moves · C4.21.4 When the anchor is recomputed—at engagement or by continuous following—affects consistency
• Adjacent: C4.20 Interaction volume versus capture range · C4.22 Perceiving interaction-volume boundaries
• Search: device-fixed · stance memory · return to position