Muscles show measurable fatigue well before a user admits to feeling tired

What it is

By the time a user says "it's fine, not that tiring," the muscle doing the work has often already shown measurable physiological fatigue. This gap between subjective feeling and objective physiological state is the piece of arm-raise interaction most easily missed in usability evaluation, because it means the absence of complaints cannot be taken as evidence that load isn't accumulating.

Why it happens

The feeling of fatigue comes from a central, integrated assessment of muscle state, and gets muddied by task focus, motivation, and the fact that a hover posture produces little visible exertion — all of which make subjective report less sensitive than measuring the muscle directly. Meanwhile, while a static load is being held, local blood flow is partially restricted by sustained tension, metabolites begin accumulating, and motor-unit recruitment patterns shift — changes that show up first in the EMG signal's spectral characteristics, such as a relative drop in high-frequency content, well before the user can clearly articulate "I'm tired." Physiological fatigue accumulates first; it only reaches awareness after crossing a perceptual threshold, and that threshold itself can be pushed later by states like focus or tension.

Studying it

Collect a subjective scale (a momentary fatigue rating taken at fixed intervals) alongside surface EMG, align them on the same timeline, and compare which curve turns or crosses a set threshold first. If the EMG measure crosses its threshold earlier while the subjective rating curve stays flat, that demonstrates the lag exists — rather than assuming it does.

Where it stops holding

The lag is most pronounced in short, high-focus tasks. If the task itself gives users ample downtime to attend to bodily sensation — long gaps between operations, low cognitive load — subjective report tracks physiology more closely and the lag shrinks. The size of the lag varies a lot between individuals, so no single fixed delay value can be given; only the directional conclusion holds: subjective report cannot substitute for physiological measurement as an immediate fatigue indicator.

Applying it

• Don't treat "users didn't complain during testing" or post-hoc satisfaction ratings as evidence that a sustained arm-raise interaction is safe long-term — this kind of feedback systematically underestimates actual load.
• When designing sustained hover operations, use a fixed time cap or posture detection as the trigger for forced interruption, rather than waiting for the user to voice discomfort.
• Verification: in the target interaction, record EMG (or at least visible compensation cues like tremor or postural sag) alongside the timing of verbal or scale-based user feedback, compare which appears first, and calibrate the actually-safe duration against the physiological or behavioral signal rather than the user's self-reported tolerance.

Related

• Same group: A8.19.1 unsupported arm-raise fatigue sets in within tens of seconds · A8.19.2 fatigue accelerates jointly with elevation angle and duration · A8.19.4 short usability tests miss this problem · A8.19.5 high-frequency actions must not require overhead reach
• Nearby: A9.09 physiological measurement of load · A9.07 subjective measurement of load
• Search terms: perceived exertion · Borg CR-10 · EMG median frequency shift · subjective-physiological gap