/home/app/projects/content-engine
1ready
0check
1draft
reel · 12-15s · 2026-05-23readyfact-checkeducation

Researchers found why ADHD women pay for a meeting hours before it starts

and the cost is never the meeting

No images yet. Visual director writes the brief, then image-generation-protocol renders the PNGs via Higgsfield.
Caption28 sentences · 372w

Researchers found why ADHD women pay for a meeting hours before it starts.

The body keys up. The jaw locks. The stomach drops at 9:14am for an 11:00 call.

Adults with ADHD report significantly higher subjective stress during the anticipation phase of social-evaluative stress tasks than neurotypical controls (Raz & Leykin, Psychoneuroendocrinology, 2015). Cortisol-level findings in adult samples are mixed; the louder, more replicated finding is the felt one. The anticipation lands harder than the event.

The brain on ADHD does not distinguish well between a future event and a present one. A calendar invite at 8:47am for a 4pm call is not stored as an item on a list. It is stored as a thing already happening.

ADHD has been associated with steeper temporal discounting of future rewards (Jackson & MacKillop, meta-analysis, Biological Psychiatry: CNNI, 2016; Cohen's d ≈ 0.43) and with what Barkley described as a "temporal myopia": a relative blindness to the future that returns behaviour to the present moment (Barkley, J Dev Behav Pediatr, 1997).

So the cost arrives early. Eighteen hours of low-grade bracing for a ninety-second exchange. The email opened, half-read, closed, opened again. The 1:1 invite with no subject line that turns the afternoon into a held breath.

Adult ADHD has been associated with altered amygdala responsiveness in fear and threat paradigms, including impaired discrimination between threatening and safe social cues (Maier et al., Psychological Medicine, 2013). Adolescent samples show amygdala hyperactivation to fearful faces (Posner et al., JAACAP, 2011). The fear system fires even at the cues that turn out to be safe.

By the time the meeting starts the work is already done. Not the meeting work. The other work. The work of having been afraid of the meeting for most of a day.

This is the anticipation tax. It is paid in advance, in a currency the calendar does not show, and the receipt arrives as fatigue with no obvious source.

Many late-diagnosed women carry this load every week without language for it. They think they are bad at meetings. The literature suggests they are paying for each one twice.

If your body keys up the day before the call, save this so it has a name the next time it happens.

Rhythm legendwhy some words are highlighted

Sentences over 18 words are tagged via tooltip. Three in a row gets a soft amber underline — the brand voice prefers rhythm variation, not monotone walls of long lines.

Visual brief

Visual brief: Ad-hoc Brief, Anticipation-tax reel

Source verdict: approved/2026-05-23/anticipation-tax-reel.md Source draft: drafts/2026-05-23/anticipation-tax-reel.md Format: reel (12-15 seconds, single near-static shot, bipartite hook overlay visible from frame 1) Generated: 2026-05-23T03:00:00Z


Visual register summary

One continuous cinematic still life of a held breath. Late-afternoon warm interior, shallow depth of field, a woman at her desk who does not move while a calendar reminder glows on the laptop screen. Palette sits in the warm cream and deep brown zone of the brand identity, with a single dusty-blue glow from the laptop as the only cool note in the frame. Typography is the brand humanist serif on top, brand humanist sans on the bottom, both ink-on-warm-light, both static for the full 12 to 15 seconds. No animation, no follow-up overlays, no music. The hook reads while the body of the subject stays still.


Production-ready units

Beat 1 (single near-static shot, 0 to 12-15 seconds, one continuous take)

  • Text overlay (verbatim from approved draft, BIPARTITE, both lines visible from frame 1, NO motion, NO replacement, persists for entire 12 to 15 seconds):

    • Top line: Researchers found why ADHD women pay for a meeting hours before it starts
    • Bottom line: and the cost is never the meeting
  • Layout (9:16, 1080x1920 canvas):

    • Both lines sit in the upper-middle band of the frame, vertically centered between 18% and 52% from the top. This zone keeps the entire overlay above the IG Reels caption block (bottom ~22% of the frame) and clear of the right-side action rail (right ~14% of the frame, vertical from ~38% to ~88% from top).
    • Horizontal text safe zone: 7% inset left, 7% inset right. Effective text column is 86% of the frame width (~929px on a 1080-wide canvas). Both lines left-aligned with a soft hanging indent (no center alignment, no full justification).
    • Top line: occupies roughly 18% to 38% from top, wraps to 2 or 3 lines at the specified size.
    • Bottom line: sits below the top line with a vertical gap of ~3.5% of frame height (~67px), occupies roughly 41% to 52% from top, wraps to 1 or 2 lines at the specified size.
    • Bottom line is visually subordinate (smaller, lighter weight) but readable in under 1.5 seconds per the editor verdict.
    • Slide-number / brand-mark indicator: none on this reel. Single-shot reels do not carry a counter.
  • Background (the AI-generated or shot footage that lives behind the overlay):

    • Cinematic warm interior. Dominant tones map to the brand base palette: soft sand #E8D9B8 for the desk surface and the back wall in soft focus, warm off-white #FAF3E2 where the window light falls, deep brown #3D2818 in the shadow side of the room and the subject's silhouette. The laptop screen glow is the only cool accent, sitting tonally near the brand dusty blue #7C95A6 so it stays in-brand even though it is a practical light source in the scene.
    • Slight grain (faint film grain at roughly 4 to 6% opacity, monochrome), no LUTs that push saturation up. Color grade slightly desaturated, warm shadows, neutral highlights, no orange-and-teal contrast.
  • Typography:

    • Top line: Recoleta semi-bold (fallback Lora bold). ~72 to 78pt on a 1080x1920 canvas. Sentence case as written. Color: warm off-white #FAF3E2 with a subtle ink-tint stroke or a soft #3D2818 ink shadow at 18% opacity, 6px offset, only where the overlay crosses light areas of the frame. Letter-spacing -1% (tightened just slightly). Line-height 1.08.
    • Bottom line: Inter, medium weight (not bold, not regular). ~38 to 42pt on a 1080x1920 canvas. Sentence case as written, no italics. Color: warm off-white #FAF3E2 at 88% opacity (slightly knocked back from the top line to establish the subordinate hierarchy). Same subtle ink shadow treatment for legibility over the light window-glow area.
    • Both lines are ink-on-warm-light in spirit, but because the cinematic plate is warm-mid-tone with a bright laptop glow zone, the overlay is rendered in #FAF3E2 (the brand warm off-white) rather than #3D2818. Reason: the brightest area of the frame is the laptop screen (cool dusty-blue glow), and warm off-white text reads cleanly against both the warm shadow side of the room and the cool laptop glow zone. Deep brown text would compete with the shadow zone and disappear on the cool glow zone. The soft ink shadow at 18% opacity (#3D2818) handles the few overlap moments where the text falls across the warm off-white window highlight.
  • Image / footage content (production-ready description, what the AI-generated or shot plate looks like):

    • Frame: 9:16 vertical. Medium shot. Subject framed from mid-chest up, positioned on the right-third vertical line so the laptop sits on the left third and the calendar reminder is the rule-of-thirds focal point in the lower-left intersection of the frame, behind the overlay zone.
    • Lens: equivalent of an 85mm full-frame portrait lens. Shallow depth of field, aperture equivalent f/2.0 to f/2.2. The laptop screen reminder is in focus. The subject's hands and shoulders are in focus. The far wall, the window, and the surface beyond the laptop fall softly out of focus.
    • Lighting: soft window light from camera-left, late-afternoon angle (low, warm, raking across the desk). Key light is the window. Fill is the laptop screen glow at lower intensity, cool. No artificial lights, no overhead fluorescent, no on-camera light. Practical only.
    • Calendar reminder on the laptop screen: a clean, modern calendar app notification panel, white card on a slightly dimmed desktop background, single line of legible text: 1:1 in 30 min. The text reads 1:1 in 30 min exactly, in a default system sans-serif (San Francisco / Inter style at the OS level), dark grey on white. The notification card is positioned center-right of the laptop screen. The screen brightness is the highest luminance value in the entire frame, by ~1.5 stops, so the eye lands on the reminder first even with the overlay above.
    • Subject: woman, mid-thirties. Hands resting on the desk, one near the laptop, one closer to the body. Shoulders raised just enough to read as held tension (not exaggerated). Mouth softly closed. Eyes on the screen, not on camera. No movement: she does not blink dramatically, does not lean in, does not sigh. The only motion in the entire 12 to 15 seconds is the natural micro-motion of breath and ambient light shift.
    • Mood: held breath. Quiet. Warm room, cold internal state. The room is comfortable; the body is not.
  • Color accents (in the plate, not in the overlay): the laptop glow reads as a single dusty blue #7C95A6 accent occupying ~6 to 8% of the frame area. No other accents. No green, no terracotta in this scene. The dusty blue is appropriate per protocol (focus, science slides) and matches the cited-research register of the caption.

  • Motion: none on the overlay. None in the camera. Subject motion is limited to involuntary micro-motion only (breath, light flicker). The overlay text is a single static composite that holds for the entire duration.


Cross-unit consistency

  • Single-shot reel, so cross-unit consistency reduces to: the overlay does not change, the framing does not change, the subject does not change scene. Loop point is clean by default because the last frame and the first frame are visually identical (static camera, no subject motion). If Andrii exports for a clean loop in the IG Reels feed, set the export so the final frame matches the first frame within 1 frame of difference.
  • Brand mark: not on this reel. Single-shot education reels do not carry the brand mark in-frame; the handle in the post header is sufficient.

Wardrobe, casting, scene styling

  • Subject: woman, mid-thirties (32 to 38 reads correctly). Realistic, not styled. Not visibly glamorous, not visibly distressed. Neutral hair (loose, not styled), no statement jewelry, minimal makeup or no makeup. The face is not the focal point; the held posture is.
  • Wardrobe: soft knit sweater or a relaxed-fit cotton button-up in a warm neutral that sits inside the brand palette: oat, soft sand, cream, warm grey. Avoid pure white, pure black, red, navy, anything saturated. Sleeves rolled or pushed up to mid-forearm so the hands read clearly on the desk.
  • Desk surface: matte light wood or warm off-white painted desk. Not glossy, not high-contrast.
  • Desk objects (sparse, brand photography style, one focal object rule): the laptop is the focal object. Permitted secondary objects: a ceramic mug in a warm neutral (oat or terracotta-adjacent, no logo, half-full of something steaming-but-not-steaming-anymore reading), a small notebook closed, a pen resting on the notebook. No phone visible. No water bottle, no plant in the foreground, no headphones. Three objects total on the desk including the laptop. Generous negative space per brand photo rules.
  • Background room: soft-focus warm interior. A window camera-left (the key light source, not in the frame itself, only its light falls in). A wall in soft focus behind the subject, neutral warm tone. Optional: a single soft shape suggesting a piece of furniture (the corner of a bookshelf, a curtain edge) in deep background blur. No people in the background, no screens in the background, no clutter, no decor that pulls focus.
  • Laptop: a generic laptop, neutral grey or warm metallic, no visible brand logo (or logo turned away or removed in post). The screen content is the only legible element: a clean calendar notification reading 1:1 in 30 min exactly.
  • Room conveys: a comfortable working-from-home space. Warm, lived-in, late-afternoon. The room is fine. The internal experience is what is not fine. That tension is the point.

AI generation prompt

Video prompt (reel, model-agnostic, ONE general prompt)

A cinematic, near-static portrait of a woman in her mid-thirties sitting at a warm wooden home-office desk in late afternoon, her body holding a quiet, contained tension as a calendar notification glows on her laptop screen. The camera holds a locked-off medium shot, framed vertically, with the laptop on the left third and the woman on the right third, no camera movement at all for the full duration. She sits very still, shoulders slightly raised, breath shallow, hands resting on the desk near the laptop, eyes on the screen and not on the camera, no dialogue, no expression change, only the smallest involuntary micro-motion of breathing. The laptop screen shows a clean modern calendar notification card reading "1:1 in 30 min" in a default system sans-serif, dark text on a white card, and that notification is the brightest single element in the entire frame by a clear margin. Soft warm window light rakes in from camera-left at a low late-afternoon angle as the only key light, the laptop screen contributes a faint cool fill, and the rest of the room falls into soft warm shadow with shallow depth of field, slightly desaturated color grade, faint film grain, no on-screen text or graphics or captions added by the model. Audio is quiet ambient room tone only, faint refrigerator-hum register, no music, no voiceover, no breathing sounds emphasized. 9:16 vertical, 12 to 15 seconds.

Image prompt (singles / per-slide carousel)

Typographic overlay only, no AI image generation needed for the overlay itself. The single AI generation for this reel is the video plate above.


Designer notes

  • Overlay is added in post, not generated. The AI video prompt deliberately instructs the model to produce no on-screen text. The bipartite hook is composited in CapCut / Premiere / IG editor on top of the generated clip.
  • Both overlay lines persist for the full 12 to 15 seconds. No fade-in, no slide-in, no replacement halfway, no exit animation. If the editor app defaults to a fade, disable it. The hook lives as a single static composite.
  • Hierarchy between the two overlay lines is established by size and weight, not by color or animation. Top line is the brand display serif at ~72 to 78pt semi-bold. Bottom line is the brand sans at ~38 to 42pt medium, 88% opacity. The bottom line is subordinate but readable in under 1.5 seconds per the editor pass.
  • Text color is #FAF3E2 (brand warm off-white) for both lines, with the soft #3D2818 ink shadow at 18% opacity only as needed where overlay crosses the brightest window-light area. Reason explained in the typography section above: the brightest point in the frame is the cool laptop glow, and warm off-white reads cleanly across both warm shadow zones and the cool glow zone. Do not switch to deep brown text without re-testing legibility on the laptop-glow region.
  • IG Reels safe zones honored: overlay sits between 18% and 52% from the top, comfortably above the caption block (bottom ~22%) and clear of the right-side action rail (right ~14%, vertical ~38% to ~88%). If Andrii adds a CTA sticker or a sound credit later, both fit below the overlay without collision.
  • Calendar reminder text is literal: 1:1 in 30 min. If the AI model's first pass renders the screen text incorrectly (common on Veo / Runway / Kling), regenerate or composite a clean notification card in post. The exact string is load-bearing for the hook; it has to read 1:1 in 30 min clearly when the viewer's eye lands on the screen.
  • Tier-1 model recommendation for the video plate: Veo 3.1 first (handles still subjects, soft window light, and shallow depth of field well, plus generates the ambient room-tone audio natively). Runway Gen-4.5 second (slightly stronger on cinematic skin tones, weaker on legible UI text on the laptop screen; composite the notification in post). Kling 2.0 third (cheap iteration, weakest on legible screen text). Sora 2 viable but the held-still subject is where Veo currently leads.
  • If the model renders the laptop screen text unreadably, composite a clean calendar notification card in post over the laptop screen region. Use a screenshot of a generic modern calendar notification with the text 1:1 in 30 min substituted in. Match the screen brightness and color temperature so the composite reads as in-frame.
  • Creative latitude flagged:
    • Tight: overlay text (verbatim), overlay positioning (above caption block, clear of action rail), overlay typography (brand stack), the calendar reminder reading exactly 1:1 in 30 min, the camera being fully static, the subject not moving, the duration of 12 to 15 seconds, no music, no voiceover, bipartite layout with both lines persistent.
    • Open: exact wardrobe color within the warm neutral range, exact desk wood tone, presence or absence of the mug and notebook secondary objects, the room's secondary background details, the exact ratio of warm-shadow to window-light pixels (any version that holds the calendar reminder as the single brightest element works).
  • Asset export specs: 1080x1920 (9:16), H.264 .mp4 for IG Reels, sRGB color profile, 30fps (matches IG default; 24fps acceptable for cinematic feel but verify IG re-encoding does not introduce judder), audio track present (room tone only, even if near-silent, so IG does not auto-suggest a trending sound on upload).
  • Open questions: none. The brief is fully specified.

Visual brief written: visual-briefs/2026-05-23/anticipation-tax-reel.md