Without arbitration, multiple alarms mask each other and critical ones go unheard

What it is

Without arbitration, alarms mask one another. When sounds play simultaneously or in uncoordinated succession, discriminability drops and users retain only part of what was issued. A critical alarm is missed not because it was never sent, but because other sounds consumed it at the output layer.

Why it happens

Masking happens in the spectrum and in attention. Spectral masking makes otherwise distinct tones hard to separate; attentional masking arises because several events compete for limited processing in a short window, so a later alarm may go unregistered even when physically clear. Combined, the failure looks like "the system says it fired, the user says they never heard it"—the hardest kind of disagreement to diagnose, because logs look clean. Time pressure amplifies it: the busier the user, the less time is available to discriminate.

Studying it

Build both concurrent and dense-sequence scenarios and measure miss rates for critical alarms against a solo-playback baseline. Measures include miss rate, error type, and response delay. Record output-layer data alongside user reports, because their divergence is the phenomenon of interest. Simulating realistic speakers or reduced sample rates tests how masking is amplified under degradation.

Where it stops holding

With few, well-separated events at distinct frequencies, masking is weak and alarms survive without explicit arbitration. Focused users performing an auditory-only task discriminate better than distracted ones, so laboratory detection rates do not transfer directly to real use. Misses can also come from volume or hardware limits and must be separated from missing arbitration.

Applying it

• Route critical events through arbitration by default; never let them play uncoordinated alongside other sounds.
• Keep sounds separated in frequency and time so they are not designed into mutual masking.
• Log both actual playback and user acknowledgement, and use the gap between them to locate masking problems.
• Verification: trigger several alarms at once and reproduce on a real device speaker, then compare critical-event miss rate against solo playback; a clear increase means arbitration is missing.

Related

• Within the group: D2.10.1 Simultaneous alarms need arbitrated ordering, not overlapping playback · D2.10.3 Arbitration must tell the user that more alarms are queued
• Adjacent: D2.10.2 Higher-priority alarms should interrupt lower-priority playback · O3.05.1 Frequent warnings lead to broad disregard
• Search terms: auditory masking · missed alarm · alarm collision