Suprathreshold latency directly causes discomfort

What it is

Once latency crosses the line at which it can be detected, it is no longer a “slightly draggy” image-quality issue. A nod, the room catches up, and the stomach answers before the rating sheet. Suprathreshold motion-to-photon latency is itself a nauseogenic stimulus. It does not need extra acceleration, turning, or story piled on. People may not say the word “latency”; they report wobble, drift, wanting the headset off.

The detection threshold and the discomfort threshold are not the same line. Just being able to see the world slide need not nauseate at once. Add more, and discomfort can arrive in tens of seconds, without filling the content.

Why it happens

Suprathreshold means the visual stable reference is already clearly fighting the head. The vestibule reported an angle, the world on the retina is late by that angle, and the error is encoded as “the environment is moving.” The environment is not supposed to move — a conflict in the same family as watching the side window from a car, except here the “car” is one’s own head. The conflict does not go through understanding: knowing that the display is late does not switch off the vestibular–visual alarm.

The faster the head move, the larger the spatial error for the same millisecond count. So suprathreshold is not a fixed millisecond number; it is whether delay times angular velocity is large enough to be taken as world motion. A nod, turning toward a sudden sound, a startle turn, are high-velocity samples. If those actions exist in the experience, suprathreshold latency is already a discomfort source — it does not wait for a latency-detection experiment.

Studying it

Set several delay steps above detection threshold and collect discomfort in short sessions (a few minutes), so it is not mixed with long-session sickness accumulation. The task must force fast head motion: acquiring a suddenly appearing target, nodding on a beat. The control is the same content with delay pushed back under threshold.

Independent variables: milliseconds above threshold, peak head angular velocity, near visual structure. Dependent variables: mid-session nausea/dizziness ratings, time to voluntary abort, after-effects on removal, whether the cause can be named as “the world chasing the head” (most people cannot, so cause coding depends on the condition contrast, not on speech).

Do not replace mid-session samples with an end total. Discomfort from suprathreshold latency can arrive fast, and the number at the end may already mix in fatigue and heat.

Where it stops holding

Subthreshold latency can still sicken through other channels in a long session; that is not this “suprathreshold directly causes” mechanism, and a long-session total should not be read backwards as it. People highly sensitive to delay (vestibular migraine, prior severe sickness) have a lower detection threshold, so their “supra” arrives earlier; a product cannot claim safety from a lab average threshold. Fully static viewing barely fires this path; using that viewing to prove “our latency is fine” is dodging the probe. Discomfort reports also move with expectation: people warned they might be sick score higher, so instructions have to be fixed. Drugs, meals, and sleep shift the discomfort threshold; repeated measures have to control these or the suprathreshold effect is measured as noise.

Applying it

• Treat fast head motion as a required test, not as an optional stress case. Calibration, the tutorial, the first minute, already have people nodding to look at their hands — those moves have to complete under threshold.
• Do not dodge suprathreshold by asking people to turn slowly. Those who can will; those who cannot (startle, social, hunting a sound) will sample the delay at a moment you did not design.
• When mid-session reports say “wobble, I want to stop,” contrast the latency condition before changing scene motion. If latency is supra, changing the scene is treating the wrong layer.
• How to check: the same thirty-second nod-and-acquire task, under threshold and at one clearly supra step. If the supra group reports wanting to stop or gastric symptoms earlier, discomfort is being explained by latency. If both groups match, the discomfort is from somewhere else — do not pin it on delay.

Related

• Same group: N1.05.1 Latency from head motion to photon update must be extremely low · N1.05.3 This latency cannot be covered up with animation
• Nearby: N1.11 Render Latency and Motion-to-Photon Time · N1.12 Refresh Rate and Persistence
• Search terms: suprathreshold MTP latency · latency-induced discomfort · visual-vestibular lag