Sometimes life is full of ambiguity, and nobody needs telling. The scan results due Thursday. The message that was read and not answered. The job you are fairly sure you did not get, and cannot yet be certain about.
What is less obvious is that this state has been measured. Ambiguity has a price, it shows up in the body, and in one experiment people were watched paying it in a currency that makes the size of the bill unusually easy to read.
Where the stress actually peaks
A team at University College London wired 45 people to electrodes and gave them a guessing game. Two rocks on a screen, and under one of them, sometimes, a snake. Guess which. When the snake turned up, so did an electric shock.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
The trick was in the rules. Every 26 to 38 trials, without announcement, the researchers changed which rock was the dangerous one, and how reliably. Sometimes the odds were heavily loaded, around nine to one. Sometimes they were a coin flip.
Across 320 trials they tracked subjective stress, pupil diameter and skin conductance, and all three followed the same quantity. Not how likely a shock was. How unpredictable it was.
Their own sentence is precise about where the worst of it sat. Irreducible uncertainty was highest, they write, when a participant’s estimated probability of a shock was 50 percent, which corresponds to a situation where the environment is utterly unpredictable. The relationship takes the shape of an inverted U. Stress rose toward the middle and fell away at both ends.
A model that used nothing but the current level of irreducible uncertainty predicted people’s stress ratings better than the alternatives. Not the shocks. The not knowing.
The claim that is not in the paper
This study travels under a punchier headline than it earns. You will see it quoted as proof that uncertainty is more stressful than knowing something bad is definitely coming.
That sentence is not in the paper. The full text was searched for it, and for any direct comparison between the 50/50 point and certainty of a shock, and there is none. What there is, is a curve with its peak in the middle.
The difference matters, because the version in circulation would be a finding about bad news, and the actual result is a finding about unpredictability. Those are not the same claim, and only one of them was tested.
What people will hand over to end it
Ten years earlier a different group ran something blunter. Thirty-two people were offered choices between pairs, and each pair combined two things: a voltage, set between 10 and 90 percent of that person’s own maximum tolerable shock, and a delay, running from one second up to 27.
Nine of the thirty-two made a choice worth sitting with. Offered a smaller shock later or a bigger one sooner, they took the bigger one.
The detail that gives this its edge is what the purchase did not buy. Taking the higher voltage, the authors note, did not cause the next trial to arrive any sooner. Nothing was saved except time spent waiting for that particular outcome. Those nine people were not buying a shorter session. They were buying their way out of the suspense, and paying for it in pain.
The other twenty-three wanted shorter waits as well, and would not pay in voltage to get them. So this is a minority, not a portrait of everybody. But it is a minority of ordinary volunteers, in a room, with a dial.
Where both studies stop
Both are acute and laboratory-bound, and both teams say so rather than leaving it to be noticed.
The London group state that theirs are not exhaustive metrics of stress, and that chronic stress does not necessarily behave like the acute kind they examined. That is a substantial limit on how far any of this travels into a long wait for scan results, which is exactly the situation a reader will have in mind.
In the dread study, the brain imaging was recorded while people waited rather than while they chose, and the authors are careful about pinning the pattern on dread specifically rather than on some other process that simply unfolds over time.
The result that changes the shape of it
There is one more finding in the London paper, and leaving it out would tilt the whole thing.
Sensitivity to uncertainty predicted how well people performed at the task. Both measures did it: the ratings people gave, and their pupils.
So the stress was not interference. It was the system correctly registering that the world had stopped being predictable, and the people whose systems registered it most sharply were the ones who did best at the thing in front of them. Feeling the ambiguity was part of handling it.
What is left
Ambiguity is not free. The cost is real enough to turn up in pupil diameter, and it is heaviest not when things look bad but when they could go either way, which is a fair description of most of the waiting anyone actually does.
What the evidence does not support is the conclusion people usually draw from it, that the not-knowing is the worst part and should be ended at any price. The discomfort is doing something. And a handful of people in a laboratory, given the option, reached for a higher voltage to make it stop a few seconds sooner, and got nothing else for the money.
Most of us have taken a worse answer early because waiting for a better one had become unbearable. That is the same transaction, with a different dial.