Skip to content

Motion-Triggered Harm

The inclusive-design failure in which interface animation, autoplay, parallax, or screen flash exceeds a susceptible user's physiological tolerance — a presentation channel that doubles as a harm channel, delivering an involuntary stimulus the median-calibrated design never accounted for.

Core Idea

Motion-triggered harm is the accessibility and inclusive-design failure mode in which interface animation, video autoplay, parallax scrolling, screen flash, or other presentation dynamics exceed a user's physiological or neurological tolerance, causing discomfort, injury, or seizure. The mechanism is a presentation-channel mismatch: the channel through which the interface delivers information (visible motion, screen luminance, frame rate, haptic vibration) is also a channel along which a fraction of users carry tight physiological tolerance thresholds that the design, calibrated at population-average tolerances, has not accounted for. Susceptible populations include people with vestibular disorders (who are harmed by large continuous motion and parallax), photosensitive epilepsy (harmed by flash above the WCAG-defined frequency and luminance thresholds of 2.3.1 and 2.3.2), migraine susceptibility, autism-related sensory sensitivities, and motion-sickness vulnerability. The harm is involuntary — the user is exposed to the stimulus without advance knowledge or prior opt-out — and the severity ranges from discomfort and loss of access to migraine induction and, at the worst end, seizure. The 1997 Pokémon "Cyber Soldier Porygon" broadcast episode, which triggered seizures in approximately 700 Japanese viewers, established the broadcast-harm case and drove international flash-rate standards. WCAG 2.x success criteria (2.2.2 Pause/Stop/Hide; 2.3.1 Three Flashes or Below Threshold; 2.3.2 Three Flashes; 2.3.3 Animation from Interactions) and platform-level reduced-motion settings (the prefers-reduced-motion media query) encode the primary defence — cap flash rates, honour reduced-motion preferences, provide static alternatives, and default to opt-in rather than opt-out for motion-heavy presentations — placing the remediation responsibility on the designer rather than requiring susceptible users to identify and configure around each site or application individually.

Structural Signature

Sig role-phrases:

  • the presentation channel — the medium the interface delivers information along: visible motion, screen luminance and flash rate, frame rate, haptic vibration
  • the dual-use of that channel — the same parallax, autoplay, or flash being legitimate information design and a harm vector, the field's default lens seeing only the first
  • the susceptible tail — a subpopulation carrying tighter physiological tolerance on that channel than the population-average design assumes (vestibular disorders, photosensitive epilepsy, migraine, autism-related sensory sensitivity, motion-sickness vulnerability)
  • the involuntary exposure — the user meeting the stimulus without warning or prior opt-out, so the harm cannot be charged to user choice
  • the tolerance gap — the harm living in the gap between the design envelope (median-calibrated) and the affected envelope (the lower-tolerance tail), invisible at the median
  • the threshold-crossing harm — the exposure exceeding the tail's tolerance, severity scaling from discomfort and loss of access to migraine and, on the flash channel, seizure
  • the least-heard concentration — harm landing on exactly the subpopulations with least voice in design feedback, so the inference must run forward from the physical channel rather than waiting for a complaint
  • the standards-encoded defence — fixed handles routed to channels, not per-disorder physiology: WCAG flash/luminance caps (2.3.1, 2.3.2), the prefers-reduced-motion switch, Pause/Stop/Hide (2.2.2), a static alternative, and an opt-in (not opt-out) default that orders exposure after consent

What It Is Not

  • Not user-triggered. The harm comes from the presentation itself, met without warning or prior opt-out, so it cannot be charged to user choice the way a clicked link could. The involuntariness is what places the fault with the designer who can fix it once, not with the exposed user.
  • Not a stylistic or aesthetic choice. Motion is not free: the channel of visible movement, luminance, and flash is also a channel along which a subpopulation carries tight physiological tolerance. Honouring reduced-motion, capping flash rate, and defaulting to opt-in are load-bearing requirements with measurable exclusion and safety consequences, not taste.
  • Not a negligible edge case. The harm is invisible at median tolerance precisely because it lands on subpopulations with least voice in design feedback — and it ranges to migraine and, on the flash channel, seizure (the 1997 Pokémon broadcast triggered seizures in ~700 viewers). Its rarity in feedback reflects who is least heard, not how severe or how real it is.
  • Not the susceptible user's responsibility to configure around. The standards-based defences exist precisely so that harmed users need not identify and reconfigure each site individually; the remediation burden sits with the designer who fixes it once for all, not with the subpopulation reconfiguring per application.
  • Not a general usability annoyance. This is a threshold-crossing of a physiological tolerance — discomfort, injury, seizure — not a matter of motion being distracting or unpleasant at the median. The harm lives in the gap between the population-average design envelope and the lower-tolerance affected envelope, not in average preference.
  • Not tolerance or equity as such. Those primes carry the substrate-neutral residue — a median-calibrated envelope a subpopulation exceeds, context-sensitive fairness to subpopulations; motion-triggered harm is the accessibility instantiation with WCAG criteria, prefers-reduced-motion, flash-rate caps, and specific susceptible-population physiology. The cross-domain lesson belongs to the parents; the standards machinery is substrate-bound.

Scope of Application

Motion-triggered harm lives across the surfaces of UX, accessibility, and inclusive design — wherever an interface delivers an involuntary presentation dynamic (motion, flash, parallax, haptics) along a channel where some subpopulation carries tighter physiological tolerance than the median design assumes; its reach is within that domain, the habitats being the media that deliver such dynamics. The cross-substrate analogues (industrial noise, fluorescent flicker, shared-space fragrance harming sensitive subpopulations) belong to the parent bundle tolerance / equity / boundary, not to "motion-triggered harm" as named.

  • Web design — parallax-scroll homepages causing vestibular symptoms, CSS animations that ignore prefers-reduced-motion, and autoplaying carousels and background video.
  • Mobile platforms — full-screen transitions producing motion sickness and haptic-feedback intensity, governed by system-level reduced-motion settings.
  • Video games and VR — motion sickness from artificial-locomotion schemes, field-of-view, and head-tracking latency, with comfort ratings and teleport-locomotion as defences.
  • Video and broadcast — photosensitive-epilepsy seizure incidents from broadcast video (the 1997 Pokémon "Porygon" episode, ~700 viewers), now constrained by flash-pattern caps in broadcast standards.
  • Public-display advertising and signage — digital billboards whose flash patterns trigger seizures or impair driver attention, subject to municipal flash-rate regulation.
  • Medical, control-room, and aviation displays — flashing alarm-attention patterns conflicting with crew photosensitivity and haptic vibration exceeding comfort tolerances.
  • Educational platforms — animated explainer videos and attention-grabbing transitions that exclude or harm a fraction of learners.

Clarity

Naming motion-triggered harm splits a single design element into two roles that a designer otherwise sees only one of. The same parallax, autoplay, or flash is simultaneously animation as information design — legitimate motion communicating change, drawing attention, signalling progress — and animation as harm vector, and the field's default lens registers only the first. The concept makes the second legible: motion is not aesthetically free, because the channel of visible movement is also a channel along which a fraction of users carry tight physiological tolerance the population-average design never encounters. Once that is named, the operative question shifts from "is this motion engaging?" to "does this presentation carry exposure for a subpopulation, and have I given them a way out?" — a question answerable against WCAG flash thresholds and reduced-motion preferences rather than against taste.

The sharpest thing the concept fixes is the locus of the involuntary exposure. Because the user meets the stimulus without warning or prior opt-out, the harm cannot be charged to user choice the way a clicked link could be, which is what reframes honouring reduced-motion settings, capping flash rates, and defaulting to opt-in from stylistic niceties into load-bearing requirements with measurable exclusion and safety consequences. It also relocates the remediation burden: the standards-based defences exist precisely so that susceptible users need not identify and reconfigure around each site individually, placing responsibility on the designer who can fix it once rather than on the harmed subpopulation that has least voice in design feedback.

Manages Complexity

The space the designer would otherwise have to reason about is a daunting cross-product: many distinct susceptible populations — vestibular disorders, photosensitive epilepsy, migraine, autism-related sensory sensitivities, motion-sickness vulnerability — each with its own physiology, crossed with many presentation dynamics — parallax, autoplay, screen flash, full-screen transitions, haptic vibration — each capable of harm along a different channel. Enumerating that grid disorder by disorder and effect by effect is unbounded, and most of those subpopulations have least voice in design feedback, so the harm is invisible at population-average tolerances. Motion-triggered harm compresses the grid by collapsing every cell to one structural question — does this presentation deliver an involuntary stimulus along a channel where some subpopulation carries tighter tolerance than the population-average design assumes? — and then routing the answer not to per-disorder knowledge but to a small set of fixed, standards-encoded handles. The designer tracks those handles rather than the physiology: the WCAG flash-rate and luminance thresholds (2.3.1, 2.3.2) give a numeric line for the seizure channel; the prefers-reduced-motion media query gives a single honour-or-ignore switch covering the vestibular and motion-sickness channels at once; Pause/Stop/Hide (2.2.2) and Animation from Interactions (2.3.3) give bounded controls for autoplay and interaction motion. The default rule collapses the remaining decision further — cap flash, honour reduced-motion, default to opt-in, supply a static alternative — so that designing safely becomes a checklist against thresholds and a media query rather than a per-user clinical assessment. The branch structure is correspondingly simple: a presentation that respects the thresholds and honours the preference is cleared; one that does not carries exposure, and because the exposure is involuntary the fault sits with the designer who can fix it once for all susceptible users, not with the harmed subpopulation reconfiguring around each site. An open, high-dimensional physiology-by-stimulus problem thus reduces to a fixed set of numeric caps, one preference switch, and an opt-in default.

Abstract Reasoning

Motion-triggered harm licenses reasoning that turns on two features the population-average design never registers: the presentation channel doubles as a harm channel, and the exposure is involuntary — met without warning or prior opt-out.

Diagnostic, on channel and subpopulation. The signature inference, faced with a design element, asks not "is this motion engaging?" but "does this presentation deliver a stimulus along a channel where some subpopulation carries tighter tolerance than the population-average design assumes?" The analyst reasons from the channel — visible continuous motion, screen luminance and flash rate, frame rate, haptic vibration — to the susceptible population it loads: large parallax and full-screen transitions implicate vestibular disorders and motion-sickness vulnerability; flash above the WCAG frequency and luminance thresholds implicates photosensitive epilepsy; flicker and harsh transition implicate migraine and autism-related sensory sensitivity. The harm is invisible at median tolerance precisely because those subpopulations have least voice in design feedback, so the inference works forward from the physical channel rather than waiting for a complaint that the harmed users are least positioned to file.

Interventionist, routing to standards-encoded handles rather than per-disorder physiology. Because the harm is a channel mismatch, the licensed intervention is not clinical assessment of each disorder but a small set of fixed handles, each predicted to neutralize a channel: cap flash rate and luminance to the WCAG numeric line (2.3.1, 2.3.2) closes the seizure channel; honour the prefers-reduced-motion media query closes the vestibular and motion-sickness channels at once with a single switch; Pause/Stop/Hide (2.2.2) and Animation from Interactions (2.3.3) bound autoplay and interaction motion; supplying a static alternative removes exposure entirely for those who need it. The analyst reasons from which channel a presentation loads to which handle disarms it, and predicts that respecting the threshold and honouring the preference converts a harm-carrying element into a cleared one — designing safely becomes a checklist against numeric caps and a media query, not a per-user diagnosis.

Boundary-drawing, on fault and on the design-versus-affected envelope. Two boundaries are drawn. The fault boundary follows from involuntariness: because the user meets the stimulus without the chance to opt out, the harm cannot be charged to user choice the way a clicked link could, so responsibility sits with the designer who can fix it once for all susceptible users rather than with the harmed subpopulation reconfiguring around each site individually. The tolerance boundary separates the envelope the design was calibrated for (population-average) from the envelope actually affected (the tail with tighter physiological tolerance), and locates the harm in the gap between them — so a presentation that respects the thresholds and honours the preference is cleared, while one that does not carries exposure regardless of how engaging it is.

Predictive, on who is harmed and on the opt-in default. The channel-mismatch frame predicts that harm concentrates in the lower-tolerance tail of each susceptible population while remaining wholly invisible at the median, so the analyst predicts that a motion-heavy default will exclude or injure exactly the users least represented in feedback — which is the reasoning behind defaulting to opt-in rather than opt-out for motion-heavy presentations: opt-out exposes the susceptible user once before they can escape, opt-in does not, so the order of exposure relative to consent is itself the design lever. The severity is predicted to scale with the channel and the susceptibility, ranging from discomfort and loss of access at one end to migraine induction and, on the flash channel for photosensitive users, seizure at the other.

Knowledge Transfer

Within UX, accessibility, and inclusive design the concept transfers as mechanism, because its load-bearing apparatus — the presentation-channel-doubles-as-harm-channel diagnosis, the forward inference from physical channel to susceptible population, the involuntariness-locates-fault-with-the-designer rule, the tolerance boundary (design envelope versus affected envelope), and the standards-encoded handle set — presupposes only an interface delivering an involuntary presentation dynamic along a channel where some subpopulation carries tighter tolerance. That holds across the domain's surfaces without translation: parallax and autoplay on the web, full-screen transitions and haptics on mobile, artificial-locomotion and head-tracking latency in games and VR, flash patterns in video and broadcast (the 1997 Pokémon "Porygon" episode), digital billboards and signage, flashing alarm patterns on medical and control-room displays, and animated educational media. The same fixed handles apply throughout — WCAG flash-rate and luminance thresholds (2.3.1, 2.3.2) for the seizure channel, the prefers-reduced-motion media query for the vestibular and motion-sickness channels, Pause/Stop/Hide (2.2.2) and Animation from Interactions (2.3.3) for autoplay and interaction motion, a static alternative to remove exposure — and the opt-in-not-opt-out default and the exclusion-of-the-least-heard prediction carry across all of them. The vocabulary does not strain across these because they are one substrate: interfaces delivering involuntary presentation dynamics.

Beyond accessibility the honest account is a shared abstract mechanism — case (B), and a well-attested one. Strip the UI idiom and a substrate-neutral structure remains: a delivered stimulus that is involuntary (the recipient cannot opt out without prior control), continuously distributed in intensity, and calibrated at population-average tolerance encounters a subpopulation in the lower-tolerance tail and harms it, in a gap invisible at the median. That structure genuinely recurs across distinct substrates as co-instances, not resemblances — industrial noise harming workers with tinnitus or noise sensitivity; fluorescent flicker triggering migraine in flicker-sensitive people; fragrance in shared offices and hospitals harming the chemically sensitive (hence perfume bans); HVAC set to median comfort excluding the lower-tolerance tail; building and vehicle vibration causing motion sickness or exceeding whole-body-vibration limits; occupational chemical-exposure limits set at the median that still harm sensitive workers; air-quality standards that injure asthmatics below the regulatory line. In each, the general pattern — median-calibrated design exceeding a tighter-tolerance subpopulation's limit along an involuntary channel — is the thing that travels and carries a real, transferable lesson: locate the harm in the gap between the design envelope and the affected envelope, place responsibility with the designer who can fix it once, and order exposure relative to consent. What does not travel is the accessibility cargo: WCAG criteria, prefers-reduced-motion, flash-rate caps, and the specific susceptible-population physiology are inclusive-design furniture that does not survive extraction. And here the residue is carried by a small cluster already in the catalog: tolerance (the explicit envelope of allowable variation a subpopulation may exceed, the nearest parent), equity (context-sensitive fairness to subpopulations), boundary (design envelope versus affected envelope), and signaling. So when the cross-domain lesson is wanted, the honest move is to carry that bundle of parents (of which motion-triggered harm is the accessibility instantiation, sibling to microcopy_ambiguity and mode_error), not the term "motion-triggered harm," whose machinery is substrate-bound (see Structural Core vs. Domain Accent).

Examples

Canonical

The defining case is the "Dennō Senshi Porygon" episode of Pokémon, broadcast in Japan on December 16, 1997. Near the episode's climax an explosion was rendered as a rapid red-and-blue strobe filling the screen at roughly 12 flashes per second for several seconds. That flash pattern crossed the frequency-and-luminance threshold that provokes photosensitive seizures, and roughly 700 viewers — mostly children watching at close range on bright television sets — were taken to hospitals with seizures, convulsions, and other symptoms. The broadcaster pulled the episode and it never re-aired. The incident directly drove the international flash-rate standards later codified in broadcast guidelines and in WCAG's three-flashes criteria, establishing that a presentation channel meant to entertain had delivered an involuntary, injurious stimulus to a susceptible subpopulation.

Mapped back: The strobing explosion is the presentation channel (screen luminance and flash rate) whose dual-use turned entertainment into a harm vector. Photosensitive-epileptic viewers are the susceptible tail; children who could not anticipate the flash are the involuntary exposure. Seizure in ~700 viewers while most were unaffected is the threshold-crossing harm concentrated in the least-heard tail.

Applied / In Practice

The prefers-reduced-motion CSS media query is the field's standardized defense, honored across modern browsers and operating systems since around 2017-2019. A user who enables "Reduce Motion" in their OS accessibility settings (iOS, macOS, Windows, Android) sets a signal the browser exposes; a designer writes CSS that reads it — @media (prefers-reduced-motion: reduce) — and under that branch disables parallax, replaces autoplaying animation with a static frame, and shortens or removes large transitions. A person with a vestibular disorder who would otherwise get dizziness and nausea from a parallax-heavy homepage sets the preference once, system-wide, and every compliant site quietly serves them the calm variant. The burden sits with the designer, who authors the reduced-motion branch once, not with the susceptible user reconfiguring each site.

Mapped back: Parallax and large transitions are the presentation channel; vestibular-disorder users are the susceptible tail meeting them as involuntary exposure. The prefers-reduced-motion switch is the standards-encoded defence — one handle closing the vestibular and motion-sickness channels at once — and authoring the reduced branch once places fault-fixing with the designer, honoring the least-heard concentration.

Structural Tensions

T1: Information design versus harm vector (the same motion, two roles). The identical parallax, autoplay, or flash is at once legitimate information design — signalling change, drawing attention, showing progress — and a harm vector, and the field's default lens registers only the first. There is no separate "good" and "bad" motion to pry apart: the very properties that make motion communicative (salience, movement, luminance change, speed) are the properties that load the harm channel, so one cannot strip the harm without dulling the signal. The tension is that motion is not aesthetically free, yet the safe move (remove it) can cost the communication it was carrying, forcing the designer to find a static encoding that says the same thing rather than simply deleting the element. Diagnostic: Does removing or reducing this motion cost information the interface must convey, and if so, is there a static or reduced encoding that carries the same meaning without loading the harm channel?

T2: Median calibration versus the affected tail (the gap where the harm hides). The design is calibrated at population-average tolerance and so never encounters the lower-tolerance tail; the harm lives entirely in the gap between the design envelope and the affected envelope, invisible at the median. This is compounded by a feedback failure: the harmed subpopulations have the least voice in design feedback, so the ordinary ship-listen-fix loop that works for median usability structurally cannot surface this harm — waiting for a complaint waits on the users least able to file one. The tension is that the standard user-centered iteration is blind precisely here, and the only reliable method is to reason forward from the physical channel to the susceptible population rather than backward from reports. Diagnostic: Is this design being validated against median feedback, or against the physical channel thresholds and a tail that will never appear in the feedback data?

T3: Involuntary exposure versus user agency (where the fault sits). Because the user meets the stimulus without warning or prior opt-out, the harm cannot be charged to user choice the way a clicked link could, and the fault sits with the designer who can fix it once. This cuts against the ordinary UX instinct to empower users with controls: a settings toggle the user must first discover does nothing for someone who is harmed on first exposure, before they ever reach the settings. The tension is that agency-based defences presuppose the user acts before the stimulus, while involuntariness means the stimulus arrives first — so the order of exposure relative to consent, not the mere existence of a control, is what determines whether the defence works. Diagnostic: Does the susceptible user meet the stimulus before or after they have any chance to opt out of it?

T4: Opt-in default versus opt-out convenience (protecting the tail by taxing the median). The involuntariness argument drives toward an opt-in default for motion-heavy presentations, because opt-out exposes the susceptible user once before they can escape while opt-in never does. But opt-in is not free: it makes the motion-rich experience the exception every user must actively enable, dulling the default for the median majority who wanted the motion and whom stakeholders design for. The tension is that the order of exposure relative to consent is the design lever, and it trades directly — opt-in is the only default that never harms involuntarily, yet it imposes friction and a flatter baseline on everyone to protect a tail that is invisible in the metrics driving the decision. Diagnostic: Does the default expose the susceptible user before consent, or withhold motion until consent at the cost of the median experience?

T5: Standards-encoded handles versus per-disorder physiology (the proxy's power and its blind spots). The great compression of the concept is routing remediation to a small fixed handle set — WCAG flash and luminance caps, prefers-reduced-motion, Pause/Stop/Hide, a static alternative — so a designer acts without a clinical assessment of vestibular disorders, epilepsy, migraine, or sensory sensitivity. But the handles are proxies for physiology, and their coverage is not total: a migraine trigger below the flash threshold, or a sensory sensitivity no media query encodes, falls in a gap where the checklist passes while a real harm remains. The tension is that the handles' power is exactly that they abstract away the physiology, and their limit is that any harm not mapped to an encoded channel or numeric line is invisible to a compliant design. Diagnostic: Is this harm channel one the standards actually encode a handle for, or does it fall in a gap that passing the checklist would not catch?

T6: One frame spanning discomfort to seizure (resisting both collapses). The concept files a spectrum under a single diagnosis: the threshold-crossing harm ranges from mild dizziness and loss of access at one end to migraine induction and, on the flash channel for photosensitive users, hospitalizing seizure at the other. Unifying them is analytically clean but operationally hazardous in two opposite directions — treat every animation as a seizure-grade emergency and legitimate motion is stripped from interfaces wholesale; dismiss the class as mere discomfort and the seizure channel is under-protected, as ~700 viewers of the 1997 Pokémon broadcast discovered. The tension is that the same frame must hold a taste-adjacent annoyance and a genuine physiological injury without flattening the difference, calibrating the response to which channel and which susceptibility are actually in play. Diagnostic: Does this channel's threshold-crossing top out at discomfort and lost access, or does it reach injury and seizure that warrant a hard numeric cap?

T7: Autonomy versus reduction (its own named accessibility failure or the instance of tolerance and equity). "Motion-triggered harm" carries proprietary machinery — WCAG success criteria, prefers-reduced-motion, flash-rate and luminance caps, and the specific susceptible-population physiology — and inside UX and inclusive design it transfers intact across web, mobile, games, VR, broadcast, and signage, because all are one substrate: interfaces delivering involuntary presentation dynamics. But beyond accessibility the named thing does not travel; what carries is a bundle of parents it instantiates — tolerance (the median-calibrated envelope a subpopulation exceeds, the nearest parent), equity (context-sensitive fairness to subpopulations), and boundary (design envelope versus affected envelope). Industrial noise harming the noise-sensitive, fluorescent flicker triggering migraine, and shared-space fragrance are co-instances of that bundle, not of the WCAG machinery. Diagnostic: Resolve toward the parents (tolerance, equity, boundary) when asking what travels beyond interface design; toward motion-triggered harm when diagnosing an involuntary presentation dynamic against a susceptible tail in situ.

Structural–Framed Character

Motion-triggered harm sits at the framed-leaning position on the structural–framed spectrum — a genuine involuntary physiological mechanism wrapped in a normative, standards-constituted inclusive-design construct. The criteria mostly point framed, with one structural thread. On evaluative_weight it leans framed: the entry names a harm and a failure mode, categories that carry a built-in verdict (an outcome to be avoided, a responsibility located with the designer) rather than a neutral mechanism the way "feedback" does — though the badness is partly substrate-grounded, since a seizure is an injury independent of any design norm. On human_practice_bound it is high: "motion-triggered harm" as a diagnosis is constituted by the practice of interface design — it presupposes a designer, an interface, a median-calibrated envelope that could have been otherwise, and the fault-location that follows from involuntariness; strip the design practice away and a strobe still triggers a seizure as brute physiology, but there is no design failure left to name. Institutional_origin is pronounced: the remediation apparatus — WCAG success criteria (2.2.2, 2.3.1, 2.3.2, 2.3.3), the prefers-reduced-motion media query, flash-rate and luminance caps, the opt-in default — is W3C and platform furniture, an artifact of the accessibility tradition, not a fact of nature (the susceptible physiology is natural; the encoded handles are not). On vocab_travels the named term scores low: its operative vocabulary is pinned to HCI and floats nowhere free of it. And on import_vs_recognize, the cross-substrate cases the entry itself lists — industrial noise, fluorescent flicker, shared-space fragrance — are co-instances of the parent bundle recognized in a new substrate, not "motion-triggered harm" imported; the named accessibility term travels only by analogy.

The portable structural skeleton is a median-calibrated envelope exceeded by a lower-tolerance subpopulation along an involuntary channel — a design (or exposure) calibrated at population-average tolerance meets a tail with tighter physiological limits on a channel the recipient cannot opt out of, and the harm lives in the gap invisible at the median. That skeleton is genuinely substrate-spanning and recurs as mechanism across noise, flicker, fragrance, HVAC, and vibration, which is what gives it real cross-domain lift. But it does not pull motion-triggered harm toward the structural pole, because that skeleton is exactly what the entry instantiates from its umbrella primes — tolerance (the explicit envelope a subpopulation exceeds, the nearest parent), with equity (context-sensitive fairness to subpopulations) and boundary (design envelope versus affected envelope) — not what makes "motion-triggered harm" itself travel: the cross-domain reach belongs to that bundle, while the WCAG criteria, prefers-reduced-motion, and susceptible-population physiology stay home-bound. Its character: a real, observer-free physiological harm channel diagnosed through a normative, WCAG-constituted design frame, structural only in the tolerance-envelope skeleton it borrows from its parents and leaves framed as an accessibility failure mode.

Structural Core vs. Domain Accent

This is the section that decides why motion-triggered harm is a domain-specific abstraction and not a prime — and it carries the case for its domain-specificity, since the entry has no separate section for that.

What is skeletal (could lift toward a cross-domain prime). Strip the interface and a thin relational structure survives: a stimulus calibrated at population-average tolerance is delivered along a channel the recipient cannot opt out of, and where the channel meets a subpopulation in the lower-tolerance tail the exposure crosses their limit — with the harm living in the gap between the design envelope and the affected envelope, invisible at the median. The portable pieces are abstract — an envelope of allowable variation set at the median, a tail that exceeds it, an involuntary channel that orders exposure before consent, and the fault-location that follows from involuntariness. That skeleton is genuinely substrate-portable, which is exactly why it recurs as the parent primes the entry instantiates: tolerance (the explicit envelope a subpopulation exceeds, the nearest parent), equity (context-sensitive fairness to subpopulations), and boundary (design envelope versus affected envelope). But it is the core motion-triggered harm shares, not what makes it distinctive.

What is domain-bound. Everything that makes the failure motion-triggered harm in particular is inclusive-design furniture that does not survive extraction: the presentation channels of interface animation, autoplay, parallax, screen flash, and haptic vibration; the specific susceptible-population physiology (vestibular disorders, photosensitive epilepsy, migraine, autism-related sensory sensitivity, motion-sickness vulnerability); the WCAG success criteria (2.2.2 Pause/Stop/Hide, 2.3.1 and 2.3.2 flash-and-luminance caps, 2.3.3 Animation from Interactions); the prefers-reduced-motion media query; the static-alternative and opt-in-not-opt-out defaults; and the standards-encoded routing of remediation to channels rather than per-disorder clinical assessment. These are the worked vocabulary and the instruments the field actually uses. The decisive test: strip the WCAG machinery and the interface substrate and a strobe still triggers a seizure as brute physiology — the harm is real and observer-free — but there is no longer a design failure to name, no median-calibrated envelope that could-have-been-otherwise, no designer to locate the fault with. Remove the substrate and the named accessibility construct dissolves, leaving only the general channel-tolerance structure.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose cross-domain transfer is recognition of the same mechanism, not analogy. Motion-triggered harm's transfer is bimodal. Within UX, accessibility, and inclusive design the mechanism travels intact — the channel-doubles-as-harm-channel diagnosis, the forward inference from physical channel to susceptible tail, the involuntariness-locates-fault rule, the tolerance boundary, and the fixed handle set all carry without translation across web, mobile, games and VR, video and broadcast, signage, and educational media, because all are one substrate: interfaces delivering involuntary presentation dynamics. Beyond accessibility the named thing does not travel; the cross-substrate cases the entry itself lists — industrial noise harming the noise-sensitive, fluorescent flicker triggering migraine, shared-space fragrance harming the chemically sensitive, HVAC set to median comfort, occupational exposure limits — are co-instances of the parent bundle recognised in a new substrate, not "motion-triggered harm" imported. What travels there is the median-calibrated-envelope-exceeded-along-an-involuntary-channel structure and its transferable lesson (locate the harm in the gap, place the fault with the designer who can fix it once, order exposure relative to consent); the WCAG criteria, prefers-reduced-motion, flash-rate caps, and susceptible-population physiology do not. So when the cross-domain lesson is wanted, it is already carried, in more general form, by tolerance, equity, and boundary. The cross-domain reach belongs to those parents; "motion-triggered harm," as named, carries the inclusive-design machinery that should stay home.

Relationships to Other Abstractions

Local relationship map for Motion-Triggered HarmParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Motion-Triggered HarmDOMAINPrime abstraction: Threshold — is part ofThresholdPRIMEPrime abstraction: Variability — presupposesVariabilityPRIME

Current abstraction Motion-Triggered Harm Domain-specific

Parents (2) — more general patterns this builds on

  • Motion-Triggered Harm is part of Threshold Prime

    Motion-Triggered Harm contains a physiological response threshold whose crossing turns presentation dynamics from tolerable stimulus into injury or incapacitation.

  • Motion-Triggered Harm presupposes Variability Prime

    Motion-Triggered Harm presupposes population variability because the inclusive-design failure lies in calibrating to a median while a lower-threshold tail remains exposed.

Hierarchy paths (2) — routes to 2 parentless roots

Not to Be Confused With

  • General usability annoyance. Motion being distracting, gimmicky, or unpleasant at the median — a taste-and-friction complaint the ordinary user files. Motion-triggered harm is a threshold-crossing of a physiological tolerance in a lower-tolerance tail, ranging to migraine and seizure, living in the gap between the design envelope and the affected envelope. Tell: is the objection that most users find the motion irritating (usability annoyance), or that a subpopulation is physiologically injured by it while the median is unaffected (motion-triggered harm)?

  • User-triggered harm. Exposure the user brought on by their own action — clicking a link, opening a video, choosing a setting. Motion-triggered harm is involuntary: the stimulus is met without warning or prior opt-out, which is exactly what places the fault with the designer rather than the exposed user. Tell: did the user act before the stimulus and could have declined it (user-triggered), or did the presentation deliver it before any chance to opt out (motion-triggered harm)?

  • Photosensitive epilepsy (the brute physiology). The neurological condition itself — the disposition of a susceptible nervous system to seize under flash. That is observer-free physiology: a strobe triggers a seizure whether or not any interface exists. Motion-triggered harm is the design-failure diagnosis built on top of it — a median-calibrated presentation that could have been otherwise, with a designer to locate the fault with. Tell: are you naming a bodily susceptibility (the physiology), or a preventable design decision that exposed that susceptibility (motion-triggered harm)?

  • Sibling accessibility failure modes (microcopy_ambiguity, mode_error). Other inclusive-design failures that share the umbrella but load different channels — ambiguous interface text, or acting in the wrong system mode. They are co-instances of the same parent family, not the same failure: motion-triggered harm is specifically the presentation-dynamics-as-harm-channel case keyed to physiological tolerance. Tell: is the breakdown in comprehension or system state (the siblings), or in an involuntary physiological stimulus exceeding a susceptible tail's limit (motion-triggered harm)?

  • Cross-substrate co-instances (industrial noise, fluorescent flicker, shared-space fragrance). Harms where a median-calibrated exposure exceeds a lower-tolerance subpopulation along an involuntary channel — noise harming the noise-sensitive, flicker triggering migraine, fragrance harming the chemically sensitive. These are co-instances of the parent bundle recognised in a new substrate, not "motion-triggered harm" imported: none involves interface animation, WCAG criteria, or prefers-reduced-motion. Tell: is the harm channel an interface presentation dynamic (motion-triggered harm), or a non-interface exposure sharing only the median-envelope-exceeds-a-tail structure (a co-instance of the parents)?

  • The parent bundle it instantiates (tolerance, equity, boundary). The substrate-neutral patterns motion-triggered harm is the accessibility instance of — a median-calibrated envelope a subpopulation exceeds, context-sensitive fairness to subpopulations, and the design-versus-affected boundary. The cross-domain lesson (locate the harm in the gap, place the fault with the designer, order exposure relative to consent) belongs to these; the WCAG and reduced-motion machinery does not. Tell: strip the interface, the flash-rate caps, and the susceptible-population physiology and what remains is bare tolerance/equity/boundary, not motion-triggered harm. (Treated fully in earlier sections.)

Neighborhood in Abstraction Space

Motion-Triggered Harm sits in a moderately populated region (56th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Communication Channels & Modality (11 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12