Two kinds of knowing

A fire commander walks into a burning building, stands in the middle of it, and — without knowing why — orders his crew out. Seconds later the floor they had been standing on collapses. He could not have told you what he saw. He just knew.

A senior doctor looks at a patient and feels, with complete conviction, that he has the diagnosis. He is wrong, and a one-line checklist a student could run would have done better. He could not have said what made him so sure. He just knew. And he was wrong.

Both are experts in their fields — the commander who reads a fire, the doctor who reads a patient. The faculty at work in each is intuition: a feeling that arrives without explanation and presents itself as knowledge. In one it saves lives; in the other it wrecks judgement. And from the inside, the two feel exactly the same.

By the end the difference between them comes down to a single question — and a short recipe for answering it, at the close (A.1), that works on any gut call you are asked to trust.

Trust the gut, or distrust it

For decades, two camps fought over which of those stories was the real one.

On one side, Gary Klein, who built much of the field that studies how real experts decide — under pressure, in the open, with lives on the line. Naturalistic decision-making, it is called: the study of experts in the act, and of what they are reading when they get it right. His work was about the marvels: the commander, the nurse who sees an infection before the tests do, the chess master who finds the move in seconds. Trust that.

On the other, Daniel Kahneman, a Nobel laureate, whose life’s work showed how systematically the mind misjudges — how confident experts are routinely beaten by simple formulas, how the feeling of certainty comes loose from the facts. Heuristics and biases, this one is called: the study of the mental shortcuts that stand in when the situation has not been read, and the errors they breed. Distrust that.

In 2009 they sat down to write a paper together, braced for a fight. They did not get one. They agreed on almost everything, and where they began apart, they converged. What was left between them, they said, was largely a difference of attitude — a taste for the marvels on one side, for the follies on the other. They titled it, dryly, A Failure to Disagree.

The paper ends by reaching past its own subject. Kahneman and Klein close it hoping someone will bring their two traditions together — the whole study of judgement, not just the question of when a gut can be trusted. That coming-together is what a full account of judgement would need to deliver.

How great minds settle an argument

They converged by dividing the ground.

Trust the gut, they concluded, where the world is regular enough to be learned and the person has had the chance to learn it: a chessboard, a cockpit. Do not, where the world offers no stable pattern to read: the price of a stock next quarter, an election years out. And within a single person they drew a finer line — expert at this task, not at that one. The same auditor can be sound on the numbers and hopeless at spotting fraud. They thought this last point important enough to name as a rule, not an exception.

It is a careful peace, and hard-won. Its shape is a map of corners — trust here, not there.

Behind it all sits the word itself: what makes someone an expert? Their field cannot score expertise the way a chess rating scores a player, so it defines it another way: an expert is whoever their peers, by consensus, recognise as performing at the highest level — the recognition counting as expertise when it reflects real success, even success no one has measured (Kahneman & Klein, 2009). When the crew says the fire would not have spread had a particular captain been in charge, that captain is the expert. It is a definition by comparison: the best practitioners set the mark, and the rest are read against them — enough to tell the able from the mediocre and the novice, and enough, the authors note, for most of what their field studies. But this names expertise by what it leaves behind — the recognition, the results — not by what produces them. Define a word that way and it can be argued over forever, settled only a corner at a time.

Unless something changes the terms.

Words to talk with, not past

Fields argue past each other more often than they argue with each other, and usually for the same reason: the words in play were never pinned down. ‘Expert’ is a good example — everyone uses it, no two people quite mean the same thing by it, and a debate about when to trust one can run for decades without anyone noticing they are arguing about different things. Decision Free Solutions (DFS) is built to close that kind of gap for organisational vocabulary generally, not for ‘expert’ alone — so that two people using a word are talking with each other, not past each other, and can build on a definition once it is fixed rather than re-arguing it each time.

The method is exact about where its words come from. DFS starts from two first principles — that reality is causal, and that people perceive those causes unequally — and adds two things that have to be put in rather than derived from them: what someone wants to happen, a desired outcome; and a precise account of what a decision is. The first is what makes ‘expert’ definable, the word this piece works with. The second carries most of the rest.

A useful side effect of this precision is that it sometimes shows two fields have been circling the same thing from opposite sides, without either having reason to notice. That is what happens here. Read through DFS’s definitions, ‘expert’ comes down to one question: how much of what bears on an outcome does a person actually perceive? Klein’s naturalistic decision-making has spent decades studying the cases where the answer is a great deal — the marvels. Kahneman’s heuristics and biases has spent decades studying the cases where it is very little — the failures. Neither tradition needed the other to be wrong; they needed a question precise enough to show they had been asking about the same thing.

What DFS adds to their careful work, then, is not a rival theory of intuition. It is the measure underneath it — what makes someone expert in the first place, rather than only the marks expertise leaves behind.

What an expert is

Decision Free Solutions sets out to do a single thing: minimise the risk that an organisation fails to achieve its goals, by bringing whatever expertise is available to bear on it. In short, to make expertise matter. What is unusual is where DFS starts. Instead of beginning from what seems to work and building outward, it begins from two first principles, adds what cannot be derived from them, and derives the rest — which is how it arrives at exact definitions for words the field has used for decades without ever pinning them down, like expert, expertise and decision.

The first: reality is causal. Things happen because of the conditions in play and the way those conditions act on each other. Behind any outcome there is a particular set of causes that brings it about: the conditions present, and the rules by which they work — some near-certain, like the way heat and a loaded floor bring a building down, others a matter of tendency, like the way an illness usually announces itself. DFS calls this plain picture the Event Model (Verweij, 2026a). A few of its terms run through everything that follows; here they are, plainly. The conditions and rules that produce an outcome are its ‘information’, and that information is there whether or not anyone perceives it. An environment holds far more conditions and rules than any one outcome turns on; those that bear on the outcome in front of you are its ‘relevant information’. Sometimes the information for an outcome is small and plain. Sometimes it is so large and tangled that no one could take it all in — the case usually called complex, which this article returns to (see A.3 below). Either way it is there. What runs short is not the causation but the perceiving of it.

The second: people perceive those causes unequally. Faced with the same situation, one person makes out far more of what is at work than another. How much of it a person actually makes out is their perceptiveness — the quiet engine under the word “expert”. Without differences in perceptiveness, the word would mean nothing.

From these two, with one thing added, the definitions build themselves. Take someone who makes out a great deal of what is at work across a whole field — cardiology, firefighting, a supply chain. That high, broad level of perceptiveness is ‘expertise’: the thing a person has. Everything so far is description — it holds whether or not anyone wants anything to happen. What the principles do not supply sits on the other side of that line: intent, someone wanting a particular outcome here and now, which is what an organisation is for. Someone whose perceptiveness reaches everything bearing on that outcome is an ‘expert’ for it: the thing a person is — and it holds for that outcome and for others like it in kind and complexity, not for the whole field. Expertise is held broadly; being an expert is settled one outcome at a time. Not a title, not a track record, not a place in the field’s ranking: a person is an expert for this outcome and not the next.

Expertise is built out of two things. One is experience: meeting case after case of the same kind, until the pattern beneath them forms. A single event, met only once, builds almost none. The other is ‘perceptive capacity’: the stable capability that determines how fast experience turns into perceptiveness — the same exposure builds more in one person than another (Verweij, 2026b). The two are not interchangeable, but they are not independent either. Someone with low perceptive capacity can reach the same perceptiveness as someone with high capacity, given enough recurrence. But where cases are never quite the same twice, experience stops being the engine — the pattern never fully forms. Perceptive capacity is what keeps running. It is what carries a person into a case they have not met before, because it picks up more of what that case shares with the underlying pattern — even when the surface looks unlike anything seen before. When that perceptiveness fires without an explanation in hand, that is intuition.

And the word the whole argument was built on falls out alongside them. Intuition is perceptiveness with its reasons unshown — the seeing has happened, but what was seen, and how it bears on the outcome, has not been put into words, and in the moment often cannot be. The commander’s certainty is the perceptiveness; his inability to say why is the reasoning, still unspoken.

Both authors endorsed Herbert Simon’s definition of skilled intuition: it is, he wrote, ‘nothing more and nothing less than recognition’ — a cue meets a pattern laid down by experience, and the pattern delivers the answer (Simon, 1992). On the mechanism, DFS reads the same thing: recognition is what built-up perceptiveness does in the moment, before any reasons are to hand. But what the person feels is only that a pattern has fired — not whether it truly fits the case or only seems to. The recognition feels the same either way. So whether an intuition is sound is never settled by its being a recognition; that waits on whether the perceptiveness underneath reached what the outcome turned on.

The definitions are not fitted to the cases after the fact; they are what those starting points leave standing. The full derivation — from the causes that make up an event, through perceptiveness, to expertise and to risk — is laid out in two of the framework’s positioning papers (Verweij, 2026a; 2026c).

Three findings, one mechanism

Three findings from Kahneman and Klein’s paper, one at a time — with the reason each gives way.

What Kahneman and Klein called a “high-validity environment” — the place where the gut can be trusted — reads, through DFS, as one question: was there anything there to read? The heat is wrong, the flames are moving in a way they should not, the floor feels different underfoot — and a commander with experience and a high perceptive capacity reads it, and gets his team out. The signs in a patient are there too — but each patient may present as one of a kind, and the same symptom can sit on a dozen different stories, so reading them fully is far harder. What will move a single stock next quarter is not there to read at all — if it were, the price would already hold it. So an expert is possible for the fire, harder-won at the bedside, and impossible for the stock: it turns on how much there is to perceive, and how much of it anyone can.

And they defined validity as the ‘causal and statistical structure’ of the environment, with its cues there to be found in principle even when the person reading them cannot say what they are (Kahneman & Klein, 2009). That is the structure the Event Model names — the conditions, and the rules by which they work, the deterministic and the probabilistic alike. Two accounts of one ground, reached from opposite directions. Where Kahneman and Klein mark an environment high in validity, DFS reads the relevant information as present and within reach; where they mark it zero, as not there to be read or beyond anyone’s reach. The gut can be trusted only where the relevant information was within reach of perceiving.

The same definition already carries what they named fractionation — their term for expertise being task-specific. An expert is an expert for an outcome, never in general. The auditor, sound on the numbers and lost on fraud, is not one expertise fraying at its edges; he was only ever expert for the numbers. There was no whole-person expertise for the task-specificity to chip away at — it was in the word from the start.

And the point they came to — that the feeling of confidence is no guide to whether a judgement is right — follows straight from what intuition is. It withholds its reasons, so the feeling is all that surfaces; and the feeling is the same whether the seeing reached what mattered or fell short of it. A sound intuition and a hollow one are, from the inside, identical. The certainty and the seeing are two different things, and the first is no evidence of the second.

There is a reason the inside view fails here. Perceptiveness reaches outward, into the conditions and rules of the case; the feeling of certainty faces the other way, and reading it tells you nothing about how far the perceptiveness got. Kahneman and Klein draw the same line: a skilled reader can catch when a situation breaks its pattern — something here does not fit — yet no one can look in on their own judgement and see whether it is sound (Kahneman & Klein, 2009). This is how an expert can know when they do not know: not by doubting the feeling, which gives nothing up, but by perceiving, out in the case, that it does not match what they have met before. The knowing points outward, where perceptiveness works — never inward, at the certainty, which reads the same either way.

So the marvels and the failures are the same thing at its two extremes: the perceptiveness gap — the distance between what bore on the outcome and what was actually perceived — wide or narrow, open or closed. Naturalistic decision-making has been working the narrow end, where experience has built the perceptiveness; heuristics and biases the wide end, where the feeling arrives with little of it underneath. Two communities, two literatures — one quantity, read from its two ends. Studied apart for decades; one thing all along.

The record, recognition, and the expert that is neither

Klein and Kahneman identified two marks expertise leaves: a record of good outcomes and peer recognition. DFS sorts them — and the sorting is where the precision shows.

A record of outcomes is real evidence of past performance — and what it warrants, when a similar situation arrives, is the expectation of a similar outcome. That expectation is called trust, and in DFS the only trust an organisation can rely on is trust that is the extrapolation of past performances into a setting of similar kind and similar complexity. Unless trust is grounded in past performance, carried forward into a similar setting, trusting an expert — in the everyday sense of deferring to their reputation — cannot be relied upon.

Peer recognition is real too, and DFS leans on it — once it is read for what it actually registers. When colleagues say the fire would not have spread had a particular hand been in charge, they are pointing at that person’s expertise. But in work where the surface varies but the underlying causal structure recurs, what sustained good reading rests on is perceptive capacity; a stored library of past cases would not carry across the variation. So the harder and more varied the work, the more it is the perceptive capacity that peers are registering, beneath the expertise it produces. And what peers are recognising matters in practice: perceptive capacity is a key part of what determines who is best suited to lead in which situations.

DFS has built a tool to identify someone’s perceptive capacity: the Perceptive Capacity Assessment Tool (PCAT). It cannot test perceptive capacity directly. What it reads is behaviour, and what behaviour expresses is perceptiveness — how much someone is actually taking in. In work involving people, situations rarely repeat in quite the same shape twice, so built-up experience carries less of the load than it does in a technical field: perceptiveness sustained across many such situations can only have been built fast, and that is what licenses reading it as a proxy for the capacity underneath. Peers who have watched someone work, in settings formal and informal, answer questions that each target one particular, observable characteristic, grouped into four categories, each a spectrum running between named ends: decision-making, from unsubstantiated to substantiated (is reasoning made available, or do conclusions arrive asserted); causality, from control to alignment (does the person work with what conditions and people actually bring, or override them); awareness, from erratic to steady (do they foresee how situations develop, or are they repeatedly surprised); and empathy, from indifference to caring (do they perceive what the people around them are actually dealing with) (Verweij, 2026b). Structured and scored, the tool aims at the capacity itself rather than at a ranking of one practitioner against another, and adds a foundation to the judgement colleagues already make. A first set of scores sits in A.2, below, for six familiar public figures — Obama, Clinton, Biden, Bush, Musk and Trump — and reads potential for performance in situations involving people, not a verdict on what any of them has actually achieved. But perceptive capacity, however high, is a propensity, not a perception. It tells you how readily someone tends to pick up what matters in a kind of setting — not what they are seeing in the case in front of you now.

Put together, the two marks share a limit. Both speak to the person’s standing across a field — recognition registers the perceptive capacity, the record earns the trust — which is to say, both speak to expertise, the thing a person has. Neither speaks to the case in hand: whether, for this outcome, here, this person’s perceptiveness reaches everything that bears on it — whether, here, the person is an expert. That is the gap between having expertise and being an expert, and it is why a recognised hand with a flawless record can be sound on one case and wrong on the next. It is why a separate condition for when the gut can be trusted had to be named at all. The two marks, by themselves, cannot tell you whether to trust the intuition in front of you.

The two marks cannot settle the case — and in the moment, there is nothing laid out to settle it with. A gut call shows no reasoning; that is what makes it intuition. So what you weigh instead is two things. First, whether the situation is the kind that holds a pattern to read at all. Second, whether this person has built the expertise to read cases of its kind and complexity — the perceptive capacity and experience their peers have come to recognise. Neither makes the call certain. Together they are the best you have before the outcome is in.

Where there is time, that bet gives way to something firmer: the reasoning, laid out — which conditions were seen, and how they bear on the outcome, set out plainly enough that someone who is not an expert can follow. That is what shows whether the seeing reached what mattered. It is why recognising an expert takes no expertise of your own, only the reasoning made visible: the chess move the other masters would never have found alone, yet know is right the instant they see it; the riddle’s answer almost no one produces, that everyone knows the moment it is spoken.

So the sorting comes out like this. Peer recognition was pointing at the person’s expertise, and — in the hardest, ever-changing work — at the perceptive capacity driving it. The record was grounding trust — the past carried forward. The feeling of certainty was never a test of anything; it feels the same whether the seeing was full or short. And the expert was none of these: the expert is whoever’s perceptiveness reaches what bears on the outcome, here. Four things untangled from two marks and one overworked word — and the careful work is not overturned but sorted, each piece set where it belongs.

What kept the commander honest, and the doctor not

This is what explains the two we began with — the commander whose gut saved his crew, the doctor whose gut failed his patient — without reaching for character at all.

The difference is not in the men, nor in how fast the world answers them. It is in what their perceptiveness had reached, and whether more was there to be had. The commander has no time to reason, and does not need it. No two fires are alike, but the physics is the same each time — heat, fuel, draught, the way a loaded floor gives — and across a career of them his perceptive capacity has picked out which signs carry the danger. That built-up perceptiveness is his expertise: what he has, and brings to every fire. Whether it makes him expert here — whether, for this fire, his perceptiveness reaches all that bears on his crew’s survival — the moment cannot tell him, and the certainty he feels is no proof of it. It turned out that it did. He perceived what mattered for keeping them alive; and since, in those seconds, no further seeing was to be had, acting on the feeling was sound. For this event, he was expert.

That the reasons stayed wordless does not make them absent. Some of them — not all — he could recover afterward, if he worked at it, by retracing the cues he took in; and the retracing would build his perceptiveness further. It is how the nurses who sensed a sick infant before they could say how were drawn out: researchers retraced the incidents with them, cue by cue, in what is called cognitive task analysis, and recovered signs of infection that had never reached the medical literature (Kahneman & Klein, 2009). Demystifying intuition by finding the cues an expert read but could not name is the central aim of naturalistic decision-making. DFS works the same seam, and adds two things to it: it says what the cues are — the conditions and rules of the case — and it turns the retracing into a step that runs forward as well as back, drawing the reading out while there is still time to act on what it misses.

The doctor has time to reason and does not use it. Reasoning the case through to himself would not guarantee the catch — a person can be blind to the faults in his own reasoning — but it gives what he has missed a chance to surface, and holds a lone certainty up against something rather than nothing. And his patterns may not hold here: patients vary more than fires, so the same signs sit on different illnesses; the ones that fit a familiar story can crowd out the ones that point elsewhere; or the illness is one he never built a pattern for at all. His perceptiveness has not reached what matters, the feeling says nothing of the shortfall, and a check that was there to run — the one-line rule, a colleague, a minute’s thought — went untaken. That is the illusion of validity exactly: confidence while the gap stays open, and obtainable seeing left on the table.

This is not to distrust the doctor’s gut as such. Where the case does fit a pattern he has truly built, his intuition is as sound as the commander’s. It is the argument for gathering rare cases into a few specialised centres rather than scattering them: the pattern for an uncommon illness forms only where enough of those cases come together to build it.

And where no pattern can form at all — the case genuinely novel, or too tangled to read — no one can be the expert who has the answer. But what the person brings still counts for a great deal: it is perceptive capacity that carries, drawing more sight from less, and the reader of highest capacity holds the best intuition available. What it cannot do is close the gap; the honest move is to say where the seeing runs out.

Perceptiveness shows even where it falls short. Someone with expertise can tell, of a given case, that they cannot substantiate it — cannot set out which conditions bear on the outcome and how — and says so. The recognition is a sign of perceptiveness, not its absence: you have to perceive a good deal to see that your own has not reached. Kahneman and Klein count it a mark of authentic expertise.

It is one of several places they reach past the plain word — genuine experts, authentic expertise, true expertise. Read through DFS, each qualifier marks what recognition and a record cannot supply: how much the person actually perceived of what the case in front of them needed. Where their perceptiveness reaches everything the outcome turns on, the person is what DFS calls an expert.

The same account explains why a plain formula can outperform a seasoned judge in the murky cases — because a rule applies the few faint signals consistently, where a tired human will not. A formula, in these terms, is a piece of reasoning made once and frozen. It holds its value while the conditions it was built for still hold, and needs re-checking when they move — which is precisely why such systems still need a human watching the world for change. Kahneman and Klein set the same condition: among the requirements for building such a rule, that changing conditions are unlikely to make it obsolete; and once it is running, that it stay under a person who monitors both its performance and the changes around it (Kahneman & Klein, 2009). The rule reads the conditions as they were when it was made; only a person can notice when they move. When no one does — the rule still applied as the conditions move past it — DFS calls that Past Decision-Making, and marks it as a source of risk: the reasoning no longer fits the case, and the gap goes unseen until something fails. And a formula is only its plainest form. A procedure, a protocol, a checklist, a best practice — each is reasoning settled once and applied later, and each carries the same risk for as long as no one is monitoring whether its conditions still hold (Verweij, 2026e).

One thread runs through all of it. A commander who reads a fire, a doctor who misreads a patient, a rule that outdoes a judge where the signals are faint — in each, what decided was never the confidence on show but whether the relevant information had been reached, and whether more of it was there to reach at all. Certainty, reputation, and record all stand downstream of that — each worth only as much as the reaching behind it.

Two fields, one account

The paper had a larger hope behind it than settling one argument. Kahneman and Klein stand for two traditions in the study of judgement: one built on its failures — the biases, the predictable mistakes — the other on its successes, the skilled reading of the expert. Each, they admit at the close, had stayed half-blind to what the other saw: the first finds error more instructive than skill, the second flinches from the word ‘bias’. A full account of judgement needs both halves, they write. Their hope was that someone might bring the two together, onto the subject both are really about.

Decision Free Solutions proposes that account. It begins from the measure this piece has been using throughout — how much of what matters a person actually takes in, and how far it reaches what the outcome turns on (Verweij, 2026a) — and treats the marvels and the failures not as two literatures but as two ends of it.

That one measure is the root of a whole framework (Verweij, 2026c), and most of it reaches well past the gut-feeling question this article has been addressing. Situations differ in how much there is to perceive — and in how much of it anyone can. Some are clear. Some are complicated but still readable. Some are complex, with too much in play at once for anyone to take in fully. Some are chaotic, changing faster than anyone can follow. DFS gives these four a name, its regions (Verweij, 2026d) — the Complex region gets fuller treatment in A.3 — and works the same way in each: draw out what people perceive, and show the reasoning as far as it reaches. Drawing it out is a defined process, with plain rules for how it is run. One rule matters above the rest: an expert sets out their reasoning until someone whose expertise lies elsewhere can follow it. That is how a person who sees one thing and a person who sees another reach each other.

And it goes wider than the single person. The tool that reads a person’s perceptive capacity also allows for what a hierarchy does: where decisions override people on authority alone, a high-perceiver may stop showing what they see, and the tool is built so this holding-back is not mistaken for low capacity. A whole organisation can be read the same way, from how it behaves: whether its choices are substantiated or made on authority alone, whether the expertise in the room is drawn on or overridden. That shows whether it draws on what its people perceive, or lets it go to waste. Leadership, here, is not a person at the top but a role — taken up wherever it is needed, by those with the capacity to see what a situation turns on: the work of creating the conditions where what people perceive gets used. A choice whose reasoning has not been fully shown — a decision — DFS treats as carrying risk to the outcome, not a verdict to defend (Verweij, 2026e). And because conditions keep changing and no reading is ever complete, the framework keeps watch: for what no one saw, and for the day a settled rule quietly stops fitting the world.

None of this was built for the fire alone. It is a proposal for how judgement and decision work more widely, and the two traditions had each been describing a part of it. Whether it amounts to the coming-together Kahneman and Klein hoped for is a bigger claim than one article can carry. The fuller argument sits in DFS’s positioning papers, which lay out the framework itself; what is here is that framework brought to bear on the specific question Kahneman and Klein asked.

There is another word

The argument was never really about intuitive expertise. It was about a word — and what one precise definition adds to decades of careful study. Pin “expert” down, and what to trust comes clear. It starts with the setting: whether the case is one where an expert is even possible — whether there was enough there to be read, and enough of it within reach. Where there was, the feeling of certainty was still never the thing to lean on; what to lean on is the reasoning, when it is shown, and the record, when it is earned — and the record only as far as the setting stays of like kind and complexity. Push the same person into a setting unlike the one their record was built in, and it tells you far less.

This is an ordinary move in science, not a grand one. Linnaeus’s taxonomy, the periodic table, natural selection — each was a precision a field had been working without, and each let scattered observation cohere. The same move can be made more than once.

Because “expert” is not the only word organisational life runs on without defining. There is a plainer one, used ten thousand times a day in every organisation on earth, and just as undefined: decision. Give it the same treatment — a choice whose reasoning has not been fully shown — and the consequences do not stay inside an academic debate. They run into what an organisation costs to operate, how well it performs, who it passes over and discriminates against, and who it quietly harms (Verweij, 2026e; 2026f; 2026g). One precise word reshaped an old argument about expertise. The next, DFS argues, does the same for how organisations are built, and what they do to the people inside them.

—-

A.1 The recipe: can this intuition be trusted?

DFS does not ask whether to trust the gut. It asks one thing: did this person’s perceptiveness reach the relevant information for the outcome — and can you tell? Before any step, set the feeling aside. Certainty is no guide: intuition withholds its reasons, so a sound reading and a hollow one are identical from the inside.

1. Is the relevant information there to be perceived?
Ask whether the relevant information for the desired outcome can be expected to be perceived at all. At one end it is present and within reach; at the other it cannot be perceived — not there to be read, or beyond anyone’s reach. Where it cannot be perceived, no intuition closes the gap.

2. Can perceptiveness have been built for cases like this?
Ask whether the situation recurs — similar conditions across cases, so experience can build perceptiveness — or whether it is possibly unique, or too complex for all the relevant information to be perceived. Where cases recur, an expert is possible. Where the situation is unique or too complex, no one perceives all that bears on the outcome and no effort closes the gap fully; the highest perceptive capacity gives the best reading available, and the honest move is to say where perception runs out.

3. Does this person have the expertise, the perceptive capacity, or both?
Identify the two, and keep them apart. Expertise is a high level of perceptiveness across the domain — what the person has, shown by a record, and carried forward only to cases of like kind and complexity. Perceptive capacity is the rate at which perceptiveness builds from exposure: it is what forms expertise quickly as similar cases recur, and what carries where no two cases are alike. It is recognised by peers, and readable through a DFS instrument built for it. Both speak to what the person brings; neither yet tells you whether, here, their perceptiveness reached what bears on the outcome.

4. Is there time to substantiate towards the desired outcome?
If there is, do not lean on the feeling. Ask them to substantiate: which conditions they perceived, and how those bear on the outcome, set out plainly enough that a non-expert can follow. That is what shows whether perceptiveness reached what mattered. If there is no time, the feeling is the bet — backed by steps 1 to 3, and nothing more.

5. Afterwards, build the substantiation.
A wordless call can usually be retraced, cue by cue — which conditions were read, and how they bore on the outcome. Retracing tests the call and builds perceptiveness for the next one.

The trust was never in the feeling. It is in whether perceptiveness reached the relevant information — shown by the reasoning where there is time, weighed from the situation and the person where there is not.

—-

A.2 From Obama to Trump, validating the Perceptive Capacity Assessment Tool

Perceptive capacity cannot be read off directly — no instrument measures it the way a stopwatch measures speed. What can be read is behaviour. Behaviour is where perceptiveness shows — how much of a situation a person is registering — and where no two situations recur in identical form, perceptiveness held high across many of them cannot have accumulated slowly. That is what makes it stand proxy for the capacity beneath. It surfaces as a pattern across the particular, observable behavioural characteristics that appear whenever someone reasons about people and organisations. The PCAT puts twenty-eight such characteristics into a peer-scored questionnaire, across the four categories already named: decision-making, causality, awareness and empathy. One question from each, to give a feel for the level of detail involved: on decision-making, does this person say openly when something is uncertain or when they don’t know? On causality, does this person recognise what others know and make room for it? On awareness, does this person foresee how things will develop? On empathy, does this person notice when something is affecting a colleague before it becomes obvious? Each question a peer has watched enough to answer is scored 5 when the behaviour is consistently there, 0 when its opposite is consistently there, and 2 when the evidence genuinely points neither way; where there has not been the chance to observe it, the question is skipped rather than guessed at.

The instrument has been calibrated against six public figures, using publicly documented behaviour across many formal and informal settings rather than direct peer scoring. The scores below read potential for performance in situations involving people and organisations specifically — not the technical, rule-governed problems a public record of achievement might otherwise suggest someone has mastered. These are environments where an outcome rarely confirms on its own that a choice was right: feedback arrives late, if at all, and is shaped by conditions the person never controlled.

The situations that reveal whether a gut call deserves trust are the same situations a leadership role presents as a matter of course: dynamic, and similar to what came before without ever repeating it exactly. Perceptive capacity is what carries someone through situations of that kind — the same capacity a high PCAT score is evidence of. A high score is therefore not a verdict on someone’s current performance in a role; it is evidence of the potential to do well in one, in the sense DFS gives that term: a function distributed throughout an organisation, not a position held by a few (Verweij, 2026h).

Subject Decision-making Causality Awareness Empathy Total Interval
Obama 35/35 32/35 35/35 35/35 137/140 95–100%
Clinton 26/35 23/35 32/35 26/35 107/140 75–80%
Biden (pre-2017) 18/35 29/35 32/35 29/35 108/140 75–80%
Biden (post-2017) 8/35 14/35 5/35 12/35 39/140 25–30%
Bush 0/35 4/35 4/35 10/35 18/140 10–15%
Musk 0/35 0/35 4/35 0/35 4/140 0–5%
Trump 0/35 0/35 0/35 0/35 0/140 0–5%

Table: Perceptive Capacity Assessment Tool (PCAT) scores for six public figures, calibrated from the public record rather than peer scoring. Each category is scored out of 35; the total sits out of 140, converted to a percentile interval.

Because all four categories score the same underlying capacity, the scores for one subject are expected to sit close together; where they do not, the divergence is usually worth investigating.

Musk is worth a closer look, because his profile tests the instrument directly. His perceptiveness in deterministic systems — engineering, physics, manufacturing — is exceptional, and nothing here disputes that. The PCAT does not score it: the instrument is bounded to human systems specifically, on the view that reading how a rocket behaves and reading how a person or an organisation behaves are different capacities, and someone can hold one fully without the other. Musk’s record in the second domain supports the boundary rather than undermining it — at Twitter in 2022 he cut roughly half the workforce with little regard for which functions were critical, then repeated an almost identical pattern at Tesla’s Supercharger team in 2024, fired and rehired within weeks, with no visible sign the first episode had changed anything. The low score is not a verdict on his engineering ability. It shows the two capacities are separable, and that a person can have one without the other.

Biden’s split is there for a related reason — a 69-point decline across the profile, concentrated most heavily in steadiness, consistent with the effect ageing has on the active, real-time reading a fast-moving situation requires.

This is a first set of scores, not a finished case: the calibration draws on the public record rather than on peer scoring, and the fuller argument for the instrument — the full question set, the scoring logic, and the correction for hierarchical position — sits in the positioning paper this scoring is drawn from (Verweij, 2026b).

—-

A.3 Reality is causal — even in situations of complexity

DFS starts from two first principles — that reality is causal, and that people perceive it unequally — which together make what DFS calls a causal-perceptive paradigm. That reality is causal is where a reader from the complexity tradition may object. The hard cases are not in dispute — a market’s mood, a crowd that turns, an organisation remaking itself, running far past what anyone can hold. Where the readings part is on what such a case is: some in the complexity tradition read it as having no cause to find — the strong emergence position, where the situation was never causal to begin with, only a tendency. DFS reads it as caused still, the cause there but not (yet) seen. No observation settles which, and the full case is made elsewhere (Verweij, 2026d).

Where the causes are there but no one can fully perceive what bears on the outcome — not as a personal limit but as a property of the situation itself — DFS calls this the Complex region (Verweij, 2026d). The label is borrowed from Cynefin but used differently: not a discrete domain but a point on a continuous spectrum of how much relevant information anyone can perceive. What is left is the same work as anywhere, only harder: draw the scattered perception together and show the reasoning, as far as it reaches. The methods Cynefin has developed — for probing a situation to see how it answers, for sensing what is taking shape, for gathering many partial views into one — do exactly that, drawing out perceptiveness. Their power rests not on the complexity reading but on how people actually perceive, which is where a causal-perceptive framework starts too. These methods are central to it.

—-

References

Kahneman, D., & Klein, G. (2009). Conditions for Intuitive Expertise: A Failure to Disagree. American Psychologist, 64(6), 515-526.

Simon, H. A. (1992). What is an explanation of behavior? Psychological Science, 3, 150-161.

Verweij, J. (2026a). Outcomes are caused, not decided: How a framework built on causality redefines risk and expertise and lays the groundwork for organisational design. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026b). Identifying and utilising individual and organisational expertise: How structured communication, process logic, and perceptiveness assessment put into practice what definitions alone cannot [provisional title]. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026c). Decision Free Solutions — a deductive organisational framework: How precise definitions reorganise what mechanistic and complexity accounts each leave incomplete. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026d). Anthro-complexity, or causality not fully perceived? Decision Free Solutions and Cynefin [provisional title]. Decision Free Solutions (in preparation).

Verweij, J. (2026e). Decisions add unnecessary risk: What a precise definition of decision reveals — after seven decades without one. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026f). Decision-making as the structural mechanism of organisational discrimination: How unsubstantiated choices generate inequality — and why awareness cannot dismantle what architecture creates. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026g). When the organisation is the trap: What a precise definition of hierarchical decision-making reveals about the psychological harm it produces. Decision Free Solutions, Theoretical Positioning Series (in preparation).

Verweij, J. (2026h). Leadership that stays when the leader leaves: How a functional definition of the leadership role reveals the architecture organisations need to utilise expertise. Decision Free Solutions, Theoretical Positioning Series (in preparation).