Safety-critical scenes must limit overlay area
Aliases: coverage budget · safety HUD area · overlay solid angle
What it is
In driving, a surgical field, or a shop-floor lift, the visible area of the real world is itself a safety margin. Overlays may exist, but how much visual angle they may occupy is a cap, not a style choice. An overlay area limit writes “how much of the real world remains visible” as a design variable for safety-critical scenes: past that area, the overlay’s usefulness is secondary to a view that is no longer enough.
What is limited is area — not a catalogue of hazard types, and not whether the whole page should clear while moving.
Why it happens
Safety-critical work is built on still being able to see that real patch: the lane, the incision, the floor under a hook. Overlay area eats that patch’s solid angle directly. Aviation and automotive HUDs keep symbols thin, conformal, and off the central visual channel because coverage predicts detection and spatial orientation. An area cap is a budget set in advance: declare a percentage of the see-through view, or a band of the field, that overlays may not cross, then pour content into the budget. When the budget is spent, later information does not enter as a solid panel.
Area is stricter than count. Ten hairlines may spend less safety margin than one half-screen plate; one large confirmation card can spend the margin in a single beat.
Studying it
In a driving simulator, surgical teaching setup, or shop-floor reconstruction, set maximum overlay coverage as the condition (for example a cap on the central field) and measure detection of real-environment events, operational accuracy, and whether people still feel they can see the work surface. The independent variables are allowed maximum area and opacity; do not let item count stand in for area. Dependent measures: detection of environment events, safety-related operational error, measured coverage.
Aviation human factors treats a dose relation between coverage and detection, not whether the overlay looks good. Using a lab comfort score as the pass line for a safety area will overestimate the margin.
Where it stops holding
Seated document work, gallery tours, scenes with no safety-critical real operation, can relax the area cap; readability will bottleneck before safety. When a second person on site is assigned to watch the real environment, the wearer’s area budget can run fuller — that is staffing covering, not the overlay becoming safe. A fully virtual training level has no real work surface, so the object of the limit disappears. Headsets differ in field of view; the same percentage is a different real solid angle, so the budget must be calibrated per device and not copied as a cross-product integer.
Applying it
- Give safety-critical mode its own area budget: solid overlay in the central field may not pass an agreed share; overflow becomes wireframe, moves to the periphery, or queues.
- Keep large confirmations, ads, and decorative plates out of that mode’s central field.
- Measure coverage on the target headset. Do not estimate only on a layout.
- How to check: in a simulated safety-critical task, turn overlays on, measure the solid fraction of the central field, and insert a change on the real work surface. Over budget, or a change routinely missed, means area was not managed as a safety variable.