Why Your Brain Hates Uncertainty — and What It Does to Your Decisions

The worst week of waiting I ever endured wasn’t for bad news. It was for no news.

A contract that would have made my agency’s year sat with a client’s board for eleven days. No update, no timeline, just silence. And I watched myself — a grown man with three decades of studying the mind — come quietly unglued. I refreshed my inbox like a lab rat at a lever. I re-read my proposal hunting for the fatal flaw. By day eight I caught myself half-hoping they’d just say no, because a no I could work with. A no had edges.

When the yes finally arrived, the relief was strangely instructive. The eleven days of maybe had cost me more sleep, more focus and more cortisol than any actual rejection that year. Nothing bad had happened. The not-knowing was the bad thing.

I’d assumed this was a personal weakness until I found the research. It isn’t a weakness. It’s a design feature — and understanding it turns out to explain an astonishing amount of human misery and bad decision-making.

The experiment that measured “maybe”

In 2016, a team at University College London led by Archy de Berker published a study in Nature Communications that put a number on what my eleven days had felt like. Volunteers played a game: turn over rocks on a screen, some of which had snakes under them. A snake meant a mild electric shock. The researchers continuously tracked stress — skin conductance, pupil dilation, self-report — while an algorithm varied how predictable the snakes were.

The result deserves to be more famous than it is. Stress didn’t peak when a shock was certain. It peaked when the odds were around fifty-fifty — maximum uncertainty. Knowing you will definitely be shocked was measurably less stressful than a coin-flip chance of the same shock. The organism would rather face certain pain than an uncertain maybe.

There was a second finding tucked inside, and it’s the redemptive one: people whose stress tracked the uncertainty most accurately also performed better at the game. The stress wasn’t malfunction. It was information — the feeling of a brain correctly flagging where its model of the world was weakest.

Why prediction is the whole business

To see why “maybe” outranks “definitely bad” on the threat scale, you have to grasp what your brain actually does all day. It doesn’t passively receive the world. It predicts the world — constantly, compulsively — and uses incoming data mostly to check its forecasts. Prediction is how a slow, expensive organ stays ahead of a fast environment: anticipate the danger, pre-position the resources, act before the event.

Uncertainty is the failure state of that entire system. An unpredictable environment means the brain can’t pre-position anything — so it does the only rational thing available: it treats the unknown itself as a threat and holds the alarm systems on standby, indefinitely. That’s the eleven-day inbox vigil. Nothing was wrong; my brain simply couldn’t verify that nothing was wrong, and unverified is, to ancient circuitry, indistinguishable from unsafe.

Psychologists call the trait version of this intolerance of uncertainty — how much distress not-knowing causes you, and how much you’ll sacrifice to end it. The researcher R. Nicholas Carleton has argued that fear of the unknown may be the fundamental fear underneath the others, and the clinical evidence backs the claim’s seriousness: work by Michel Dugas and colleagues has shown that intolerance of uncertainty is a core engine of chronic worry — worry being, in their analysis, not a malfunction but a strategy. The worrier isn’t enjoying the spiral; they’re using mental rehearsal of every possible outcome to manufacture a feeling of preparedness. Worry is what reaching for certainty looks like when no certainty is available. If that loop owns your evenings, I’ve written a field guide to breaking it in how to stop overthinking.

The tax on your decisions

Here’s where this stops being an anxiety story and becomes a decision-making story — because a brain that hates uncertainty doesn’t just suffer. It pays to make the suffering stop, and the currency is usually decision quality. Four taxes worth knowing by name:

Premature closure. Under uncertainty, any answer feels better than no answer — so we grab the first plausible one and stop looking. Hiring the mediocre candidate because the search was unbearable. Accepting the first diagnosis, the first offer, the first explanation. The decision ends the discomfort; that’s what it was for. Whether it was right becomes tomorrow’s problem.

Paralysis and the certainty mirage. The opposite failure: refusing to move until you’re sure — one more analysis, one more opinion, one more quarter of data. But in most real decisions certainty never arrives, so “waiting until I’m sure” is just deciding not to decide while feeling responsible about it. Meanwhile the option ages, the competitor moves, the house sells to someone else.

Paying the certainty premium. People will accept objectively worse outcomes in exchange for predictability — the dead-end job kept for years over the promising leap, the known-bad relationship over the unknown alone. Behavioural economists find versions of this everywhere: we don’t maximise outcomes; we minimise maybes.

Story-grabbing. Most insidiously, an uncertainty-starved brain fills information gaps with invented certainty — usually threat-flavoured. The silent director is planning your exit. The unanswered text is a verdict. This is where intolerance of uncertainty shakes hands with the thinking errors I’ve catalogued in cognitive distortions: mind reading, fortune telling and catastrophising are all, at bottom, counterfeit certainty — the brain forging the answer it can’t obtain.

Look at the four together and notice the common structure: in every case, the decision is being optimised to end the feeling of not-knowing rather than to produce a good outcome. The discomfort has quietly replaced the goal.

Tolerance, not elimination

The self-help instinct here is to hunt for calm — to find some technique that makes uncertainty stop feeling bad. I want to argue for something more honest: it will always feel bad, because the feeling is your predictive machinery working. The trainable skill isn’t eliminating the discomfort. It’s raising your tolerance — widening the gap between feeling the uncertainty and paying to make it stop.

That reframe matters because intolerance of uncertainty behaves like a fitness level, not a fixed trait. Clinical work by Dugas’s group treats it exactly that way — successful therapy for chronic worriers targets the intolerance itself, deliberately practising exposure to not-knowing, and the worry recedes as the tolerance grows. You don’t drain the ocean. You get better at swimming.

And underneath the skill sits a belief worth interrogating directly. People with low tolerance almost always carry some version of “I cannot cope unless I know” — a limiting belief masquerading as prudence, usually installed by a period of life when unpredictability genuinely was dangerous. Beliefs like that don’t yield to argument; they yield to contrary evidence, gathered deliberately — the full method is in how to change a limiting belief. Every maybe you survive on purpose is evidence in the file.

The drill: deliberate maybes

Here’s the practice, borrowed in spirit from the clinical protocols and shrunk to civilian size. Once a day for two weeks, choose a small uncertainty on purpose — and don’t resolve it.

The menu is everywhere once you look:

• Order something unfamiliar without googling it first.
• Send the email and don’t check for a reply for four hours. Sit in the not-knowing.
• Let someone else pick the film, the route, the restaurant — and ask no questions.
• Make one small, low-stakes decision at 70% information, deliberately, and note the urge for the other 30%.
• Leave a question you’re itching to look up unanswered until tomorrow.

Two rules make it training rather than quirk. First, name the moment: when the discomfort spikes, say inwardly, this is uncertainty, not danger — my prediction engine complaining, not a threat report. You’re teaching the interpretation layer, not suppressing the signal. Second, log the outcome: each evening, one line — what was the maybe, how loud was the discomfort (1–10), what actually happened. Within a fortnight you’ll own a private dataset with an almost comically consistent finding: the discomfort averaged a six; the outcomes averaged fine. That dataset — not affirmations, not reassurance — is what recalibrates the machinery.

The prize is larger than comfort. Every genuinely valuable thing in your life sits on the far side of a maybe — the pitch, the move, the conversation, the launch. People who need certainty first don’t get certainty; they get smallness, because certainty-first is a bid your life can’t accept. The mindset work here isn’t learning to love the fog. It’s learning that you can navigate in it — and that the voice screaming for the fog to lift first was never going to be satisfied anyway.

Eleven days of silence taught me what the UCL study measured: the maybe hurts more than the no. But the maybe is where everything gets decided. Better to train for it.

---

Want a structured way to start? My free 7-day email course, The Reset, walks you through the core mechanics of change — one short, practical email a day. No woo, no false certainty. Just mechanism. Start here and join The Reset →

— David Watts