Decision-Making Under Pressure

Most of the decision methods you have met so far assume you have time: list the options, weigh the trade-offs, run the numbers. Real pressure removes that luxury. A shift lead facing a sudden outage, a nurse reading a crashing patient, a driver reacting to a swerving truck — none of them can convene a meeting. They must decide now, with partial information, while stress narrows their attention.

The good news is that people can decide well under pressure, and we know a fair amount about how. The bad news is that the same fast machinery that saves an expert can badly mislead a novice, or an expert working outside their domain. Knowing which situation you are in is the real skill.

How experts actually decide fast

When researchers watched experienced professionals make life-or-death calls, they did not find people rapidly comparing options. They found people recognising the situation. The expert sees a pattern that resembles hundreds of past cases, and that recognition brings a plausible course of action to mind already attached. This is the heart of Gary Klein's recognition-primed decision model: experience does the sorting, so the first workable option is usually good enough to act on.

Mental simulation, not comparison

The expert does not then line up rival options. Instead they run a quick mental simulation: they imagine carrying out the action and ask whether it would work. If it would, they do it. If they spot a problem, they adjust or reach for the next option that recognition offers. This is why skilled people can seem to act without deciding — the comparison happens inside a single imagined run-through, not across a table of choices.

When intuition earns trust

Intuition is not magic; it is recognition of patterns you have genuinely learned. That learning only sticks in what psychologists call high-validity environments: settings that are regular enough to contain real patterns, and that give feedback quickly and clearly enough to teach them. Firefighting, chess, and anaesthesiology qualify. Long-range political or stock-market forecasting does not — the world is too irregular and the feedback too slow and noisy. Outside a learnable, regular environment, confident intuition is often just confident guessing.

Imagine a duty manager at a distribution centre when a conveyor jams during peak dispatch. There is no time for a full analysis. A seasoned manager does not brainstorm ten options; she recognises the jam's signature — the sound, the spot, the backlog forming — as the kind she has cleared before. One action comes to mind: stop that line, reroute parcels to the second belt, send two staff to clear the jam. She pictures it for a second: will rerouting overload belt two? Probably not at this hour. She acts. Total elapsed time, perhaps fifteen seconds. That is recognition-primed decision-making in ordinary work.

Now put the same manager in a domain she has never worked in — say, judging which of two suppliers will stay solvent next year. Her gut will still produce a confident answer, because the mind generates intuitions readily. But solvency a year out is a low-validity problem: irregular, slow to give feedback, easily swayed by a firm's polish. Here her intuition is worth little, and a slow, structured comparison would beat it. Same person, same confidence, opposite reliability — because the environment changed.

Gary Klein, a research psychologist, studied how experts decide under pressure by observing and interviewing fireground commanders in the 1980s. He expected them to compare options and was surprised to find they usually did not; instead they recognised a situation and acted on the first workable plan. He developed this into the recognition-primed decision model, described in his 1998 book "Sources of Power."

Daniel Kahneman, a psychologist known for cataloguing the failures of intuitive judgment, took the opposite starting point. Rather than argue past each other, the two ran what they called an adversarial collaboration and published "Conditions for Intuitive Expertise: A Failure to Disagree" in the journal American Psychologist in 2009. Their shared conclusion is the practical takeaway of this lesson: expert intuition can be trusted only when the environment is regular enough to hold real patterns and offers feedback fast enough to learn them. Where those conditions fail, the same intuition becomes overconfidence.

• "Fast decisions are careless decisions." Trained experts decide fast by recognising patterns, not by skipping thought.
• "Experts weigh all the options." Under pressure they usually act on the first workable option and simulate it forward.
• "Intuition is reliable if it feels certain." Certainty tracks fluency, not accuracy; low-validity fields produce confident errors.
• "Slow analysis always beats gut feeling." In regular, well-practised domains, expert intuition can match or beat slow analysis.

You will be shown a series of high-pressure decisions across different fields — an emergency room, a trading floor, a kitchen, a boardroom — and asked to judge whether a snap intuitive call should be trusted or checked, based on how regular the environment is and how fast its feedback arrives.

Think Like a Maester: Under pressure, ask not "do I feel sure?" but "is this the kind of situation my gut has actually had the chance to learn?"

1. In the recognition-primed decision model, what does an expert do instead of comparing options?
2. What is mental simulation, and where does it fit in a fast decision?
3. Name the two conditions an environment must meet for intuition to be trustworthy.
4. Why can a genuine expert's intuition fail badly in an unfamiliar domain?
5. What did Klein and Kahneman agree on in their 2009 paper?

Under time pressure, good decisions rarely come from weighing every option. Experts recognise a familiar pattern, retrieve a plausible action, simulate it briefly, and act. That machinery is reliable only in high-validity environments — regular enough to contain patterns and quick enough to teach them. When you face a fast call, the key question is not how confident you feel but whether your experience has truly had the chance to learn this kind of situation. If it has, trust the recognition and move. If it has not, slow down and lean on structure, because a confident gut in an unlearnable field is just a guess in a hurry.

• Decision Frameworks
• Second-Order Thinking