Every focus app now has an AI coach. A sober look at what generative AI can do for your attention - and what is just a chatbot in a calm font. Read →
Can you score brain fog? Turning a vague feeling into a number
Brain fog is the word we reach for when thinking feels like wading. It refuses to sit still long enough to be measured. But the things that tend to produce it - short sleep, too much switching, too little recovery - leave a trail you can count. You cannot score the fog. You can score its inputs, and let the number stand in as an honest estimate.
Almost everyone has had a day where the lights are on but nothing connects: you reread the same paragraph, lose the thread of a sentence halfway through, and feel slow in a way that has nothing to do with how much you care. We call it brain fog. It feels like a property of your brain, fixed and mysterious. More often it is the visible end of a few ordinary, measurable inputs.
What brain fog is
Strip away the mystique and brain fog is mostly attention that is under-slept, over-switched and under-recovered. Under-slept: you did not get enough hours, or you got them at the wrong time, so the maintenance work sleep does went unfinished. Over-switched: your attention was fragmented across dozens of tasks and pings, never settling long enough to do anything deep. Under-recovered: you sat still, skipped movement and never gave your system a chance to reset between bouts of effort.
None of those are diseases. They are states, and states change. That is the hopeful part: if fog is the sum of a few inputs, then the inputs are levers, not verdicts.
The inputs you can measure
Three families of signal do most of the explaining, and a phone already sees all of them.
- Sleep duration & timing. Not just how many hours, but whether they landed at a consistent time. A late, short, ragged night shows up in the morning as fog far more reliably than a single early alarm.
- Focus continuity. How long your attention held before it scattered. Ten minutes of unbroken deep work is a different signal from ten minutes chopped into thirty fragments.
- Recovery & movement. Whether the day had any physical reset in it. Movement is one of the most boringly effective things you can do for clarity, and it is trivial to count.
Combining them into one daily clarity score
Once you have three measurable inputs, the obvious move is to fold them into a single 0-100 number you can read each morning - call it a clarity score, the inverse of fog. The recipe is not magic; it is a weighting. A reasonable, illustrative split looks like this:
Read it the right way round and a clarity score is just a tidy summary of three things you already half-know about your day. The value is not the digit. It is that one glance makes the pattern legible: most of your foggy mornings will have a short, late night sitting right behind them.
62
Today’s clarity an estimate, not a diagnosis
Where AI helps
A bare number tells you the weather, not why. This is where a generative layer earns its place: explaining the trend. Good AI narration looks at the run of recent days and says something specific and falsifiable - your clarity dips most on mornings after you slept under six hours, or your foggiest afternoons follow days with no movement at all. That is useful, because it points at a lever you can pull tonight.
The framing matters here. The score is a reflection of inputs you can change, and the explanation is a prompt to notice a pattern. Treat both as a mirror, not a meter.
Where AI overreaches
The line is bright and worth stating plainly: a clarity score can describe your day, but it cannot diagnose you. The moment a tool moves from “your sleep was short this week” to naming a condition, it has left what the inputs can support. Fog has many causes a phone cannot see - illness, medication, stress, things that belong to a doctor, not an algorithm. A fluent explanation is not a correct one, and a number is not a medical measure. If fog is persistent or frightening, that is a conversation for a clinician, not an app.
How Soren thinks about it
Soren does not score brain fog, and it will never claim to diagnose. It computes one daily 0-100 Soren Score on your iPhone from three pillars - Focus from your deep-work sessions, Sleep auto-detected from iPhone motion and written to Apple Health, and Physical from HealthKit movement; the weights are published, so you can see how the Soren Score works. Those are the same three families of input that tend to sit behind a foggy day, which is why the score so often tracks how clear you felt.
It is private by design: no account, no servers, no analytics, nothing leaves the device. The number is offered as an honest estimate and a reflection of your day, explicitly not a clinical reading and never a diagnosis. Sleep stages in particular are modelled estimates, not lab measurements.
The weightings and scores in this piece are illustrative, chosen to show how inputs could combine into a single number - they are not measured figures or a clinical formula. Soren is iOS 17+, on the App Store. Soren is not a medical device and does not diagnose, treat or assess any condition. If brain fog is persistent, speak to a clinician.
Attention scores are not a diagnosis. Here is what an ADHD-style score can honestly tell you, what it cannot, and how to use it kindly. Read →
AI brain-health apps are multiplying fast. How to tell a useful score from a confident guess - and what to ask before you trust one. Read →