Aura isn't a warning light. It's a signature you can read.

The thing you watch and then throw away

The zigzag starts near the center of your vision and creeps outward, a shimmering arc that bends the edge of whatever you look at. You know the drill. Fifteen minutes, maybe twenty, then it fades and the head pain moves in behind it. So you wait it out, because that's the only move the moment offers.

What almost nobody does is write down what it actually looked like. Not because it doesn't matter. Because every tracker you've tried handed you a single box to tick: aura, yes or no. The most information-dense event in the whole attack, and the app reduced it to one bit.

That's a strange thing to accept when you think about it. You just watched a reproducible neurological event march across your visual field on a stopwatch, and the record of it is a checkmark.

The aura signature

Here's the frame we work from. Your aura has a signature: a modality (what sense it hits), an onset time relative to the headache, a march (the order symptoms arrive in), and a duration per symptom. Those four properties are stable enough, attack to attack, that a neurologist can often recognize a patient's pattern from a good description alone.

A binary flag captures none of it. "I had aura" and "I had aura" describe a fifteen-minute scintillating scotoma one week and a forty-minute wave of numbness climbing your arm into your lip the next. Same checkbox. Completely different neurology, and potentially a different diagnostic conversation.

The signature is the data. The flag is the shadow it casts.

The signature is the data. The flag is the shadow it casts.

Why the checkbox stuck around

The one-box design isn't malicious. It's a UX default that treats aura as a prodrome alarm: the warning that a headache is coming, useful mainly as a cue to take a triptan or a gepant early. Under that model, all you need is the alarm's on-off state. Timing? Modality? Irrelevant to the alarm.

But aura earns its own diagnostic weight independent of the pain that follows. It shows up in migraine with aura, in silent (acephalgic) migraine with no headache at all, and its specific features are how clinicians separate typical aura from the variants that need a closer look. Collapsing all of that into "did the alarm fire" optimizes for the one job aura does before an attack and ignores everything it tells you afterward.

And afterward is where you live. The aura, the headache, then the postdrome tail that most tracking pretends is over. A signature you can actually read connects those phases instead of amputating them.

What the literature actually describes

The International Classification of Headache Disorders, 3rd edition (ICHD-3) defines a typical aura as one or more fully reversible symptoms, each developing gradually over at least 5 minutes, with symptoms often occurring in succession, and each lasting 5 to 60 minutes. That timing window is a diagnostic criterion, not a vibe. A "visual disturbance" that lasts three seconds or three hours is telling you something different from a textbook aura.

The accepted mechanism is cortical spreading depression, a slow wave of neuronal activity followed by suppression that propagates across the cortex at a few millimeters per minute. That propagation speed is why the scotoma marches rather than blinking on, and why a sensory aura can crawl up a limb in a way you can time.

Visual disturbance is the most common form, but it's one of several ICHD-3-recognized modalities, and which one you get, in what order, is the part a checkbox erases.

The same attack, logged two ways

Picture two months of the same person's data.

In the binary version: aura checked on eleven days. That's the whole record. A neurologist looks at it and learns almost nothing beyond frequency.

In the signature version: on eight of those days, a visual scotoma started 30 to 40 minutes before the headache and cleared inside the hour, classic and reassuring. On three days, something else. A sensory aura that climbed the left arm, arrived with word-finding trouble, and stretched past ninety minutes. In the first record those three days are indistinguishable from the other eight. In the second, they're a pattern that just earned a conversation about whether those specific attacks are behaving like typical aura at all.

That difference isn't cosmetic. It's the difference between eleven checkmarks and a timeline a specialist can actually reason over.

What changes if aura is data

If the signature is the unit, a few things follow. Your own baseline becomes legible: you learn what your typical aura looks like, so the atypical one stands out instead of blending in. The lead time between aura onset and headache becomes a number you can act on, not a guess. And the aura stops being a severed event and becomes the opening frame of a continuous record that runs through the headache and into the postdrome hours the pain-focused tools drop.

None of that requires a new mechanism or a prediction engine. It requires capturing what the moment already contains and refusing to round it down to one bit.

How we designed for the moment you can't see straight

The hard part is obvious the second you try it: logging structured detail during an active visual aura is close to impossible when the screen is the thing shimmering. A form that demands four fields while your central vision is a fortification of zigzags is a form you'll abandon, and then you're back to the checkbox by default.

So Aura Mode in Postdrome is built for exactly that window. Large targets, minimal reading, capture the modality and the onset in a few taps, and fill in the march and duration once the aura passes and your vision returns. The record stays on device, and it exports in a form you can hand to a headache specialist rather than screenshot and apologize for. The point isn't a smarter log. It's a log you can actually complete while the event is happening, so the signature survives instead of collapsing into one tick.

What we're watching next

The open question we keep circling: how much of the postdrome tail correlates with aura modality and duration, not just headache severity. Anecdotally, longer and sensory-heavy auras seem to drag a heavier fog behind them, but that's a pattern to test against real timelines, not a claim to make. The structured record is the prerequisite. You can't study the tail if the first frame was a checkbox.