The susceptible population is not negligible

What it is

Vestibular sensitivity is not a rounding-error of users. Migraine, vestibular disorders, and a slice of older adults produce symptoms on ordinary scroll parallax and full-page scale. Treating them as “a tiny minority who can put up with it” is letting the people on the team who do not get dizzy set the default for everyone. The share is not negligible, so large optic flow cannot be patched after the fact with an in-app toggle — people who never find the toggle are already exposed.

Why it happens

Susceptibility is skewed: most people show almost nothing on a short exposure; the tail, on the same stimulus, rapidly hits nausea, imbalance or a migraine attack. Mean comfort hides the tail because scores are flattened by the majority’s “fine.” The same person also moves with state: short sleep, a migraine prodrome, just out of a car, and “fine” drops to “must stop.”

Lab recruitment systematically misses the tail: ads that say “animation study” keep sensitive people away, and those who come may endure to finish. Product logs miss them too — their behaviour is to close the site and never return, not to leave a one-star review.

Studying it

Do not rely on mean discomfort from a convenience sample. Actively recruit people with migraine or vestibular history, and report mid-session dropout instead of deleting it as missing. Use a high quantile (the top tenth, for example) rather than the mean. Screening questionnaires help but do not replace exposure: people who say they are “usually fine” may still terminate early on a particular parallax page.

Where it stops holding

• The share changes by genre: games and VR start higher; a productivity product must not take game-player tolerance as the bar.
• “Nobody around me has mentioned it” is not epidemiology. Design teams are homogeneous in age and health.
• Sensitive does not mean “can never see any motion.” Small objects, low speed and skippable short transients are often acceptable. What is not acceptable is large flow as the default.

Applying it

• Default large parallax and full-page scale to off, rather than on until complaints arrive.
• The validation roster for sensitive paths must include people with relevant history, not only the team.
• Check: watch dropout and use of “turn motion off,” not mean star ratings. A healthy average with dropouts clustered on one parallax page means the tail was ignored.

Related

• Same group: F7.09.1 Large translations and scales readily cause discomfort · F7.09.2 Parallax scrolling is high-risk motion
• Nearby: N1.13 Causes of motion sickness and individual differences · F7.08.1 The OS reduced-motion setting must be honoured · A11.01 Age-related capability change
• Search terms: vestibular susceptibility · migraine · vection