Naturalistic Decision Making

Naturalistic Decision Making is a research community, also known as NDM, that studies how experts work in organizations and other consequential real-world environments. It emerged in opposition to an approach centered on laboratory demonstrations of heuristics and biases, emphasizing instead the practical intelligence of people who must act under pressure, incomplete information, and changing circumstances. Its intellectual leader in the account is Gary Klein, whose work with fireground commanders supplied the clearest model of its view of expertise.

Identity and Intellectual Position

Naturalistic Decision Making rejects the idea that the most important evidence about judgment comes from artificial experiments designed chiefly to expose failure. Its members criticize the heuristics-and-biases approach for being overly concerned with mistakes and insufficiently attentive to real people performing tasks that matter. They are also deeply skeptical of rigid algorithms intended to replace human judgment, and they do not treat Paul Meehl as an intellectual hero. The contrast is therefore not simply between two theories of intuition; it is between two research cultures, one organized around the limits of judgment and the other around the achievements of skilled practitioners.

NDM does not imply that intuition is always correct. Its central question is more discriminating: under what circumstances does an experienced professional’s immediate judgment reflect acquired expertise, and under what circumstances is it only a compelling guess? That question became the basis for a sustained exchange between Gary Klein and Daniel Kahneman, whose own work had been shaped by the illusion of validity and by demonstrations that clinical prediction could be inferior to simpler methods.

Gary Klein and the Adversarial Collaboration

The relationship between Naturalistic Decision Making and Kahneman’s research was initially adversarial. Kahneman described NDM scholars and practitioners as people who rejected his customary focus on biases, while Klein had articulated the opposing position over many years. Yet Kahneman had long admired Klein’s studies of firefighters and his analysis of how experienced professionals develop intuitive skills. Rather than allowing the disagreement to become another hostile academic exchange, Kahneman invited Klein to collaborate on mapping the boundary between the marvels and the flaws of intuition.

Their project became an adversarial collaboration: scholars who disagree about the science jointly investigate their differences, sometimes under the supervision of an arbiter. Over seven or eight years, Klein and Kahneman held many discussions, resolved numerous disagreements, nearly ended the project more than once, wrote successive drafts, became friends, and published a joint article titled “Conditions for Intuitive Expertise: A Failure to Disagree.” The title captured the result precisely. They did not encounter major issues on which they genuinely disagreed, but neither did they simply collapse their distinct intellectual positions into one.

Recognition-Primed Decision Making

Klein’s account of expertise grew from his observations of fireground commanders, the leaders of firefighting teams. He followed commanders as they fought fires and later asked how they had made their decisions. The initial hypothesis was that they would compare a pair of options and select the better one. The evidence contradicted that picture: commanders usually generated a single plausible option, considered it first, and did not need to construct an extensive menu of alternatives.

Their ability came from a repertoire of patterns accumulated through more than a decade of real and virtual experience. When a situation activated a familiar pattern, a possible course of action appeared rapidly. The commander then mentally simulated that action to test whether it would work. If the plan seemed appropriate, it was implemented. If it had shortcomings, it was modified; if it could not be modified easily, the commander moved to the next most plausible option and repeated the process.

Klein developed this account into the recognition-primed decision model. The model combines the two forms of cognition described as System 1 and System 2. System 1 uses associative memory to produce a tentative plan automatically through pattern recognition; System 2 then examines the plan deliberately by simulating its likely consequences. The model applies beyond firefighting, including to expertise in chess, and develops ideas associated with Herbert Simon, whom competing schools of decision research regard as a founding figure.

Conditions for Trustworthy Intuition

The collaboration’s principal conclusion was that intuitive expertise can usually be distinguished from bogus intuition by examining its provenance rather than the subjective force of the judgment itself. Two conditions are essential. First, the environment must contain enough regularity to be predictable. Second, the professional must have had a prolonged opportunity to learn those regularities through practice. Where both conditions hold, associative memory can recognize relevant situations and generate rapid, accurate predictions or decisions.

The same analysis explains why confidence cannot serve as a dependable test of expertise. Cognitive ease and coherence can make a story feel true because it comes readily to mind without contradiction or a competing scenario. The associative system can suppress doubt and retrieve information consistent with the currently dominant story, allowing people to feel highly confident while overlooking what they do not know. Klein and Kahneman therefore agreed on a stringent principle: the confidence people place in their intuitions is not a reliable guide to their validity, and no one, including oneself, should be trusted to determine how much trust a judgment deserves merely by reporting confidence.

NDM’s position also requires attention to the boundaries of a skill. A psychotherapist may be able to anticipate what a patient will say next without being able to predict how well that patient will be doing a year later. A financial professional may possess useful knowledge in many parts of the trade without being able to pick stocks, and a regional specialist may understand the Middle East without being able to forecast its future. Short-term anticipation and long-term forecasting are different tasks, and experience in one does not automatically transfer to the other. Experts may therefore possess genuine intuitive abilities while remaining unaware of the situations in which those abilities fail.

Broader Significance

Naturalistic Decision Making gives expert intuition a conditional defense rather than a blanket endorsement. It explains why firefighters and other experienced professionals can act effectively without comparing every available option, while also preserving a decisive distinction between learned recognition in a regular environment and confident prediction in a low-validity domain. The result is a more precise account of expertise: intuition is not validated by speed, vividness, or confidence, but by the structure of the environment and the quality of the feedback that practice has supplied.

This position stands within a wider dispute about human judgment and rationality. The behavioral tradition distinguishes real Humans from the frictionless rational agents of economic theory, showing that people are vulnerable to framing, reference points, and other systematic departures from formal consistency. Yet the existence of bias does not prove that every intuitive judgment is defective. NDM contributes the complementary warning that a critique of bias becomes incomplete when it cannot explain how skilled practitioners make good decisions in demanding settings.

The community’s importance therefore lies in the boundary it draws. It neither treats experts as infallible nor assumes that formal rules automatically outperform judgment. Instead, it asks whether the task is regular enough to teach, whether the practitioner has had sustained practice, whether feedback has made the relevant cues learnable, and whether the claimed expertise is being applied to the task actually mastered. In that sense, Naturalistic Decision Making turns the question of intuition from a contest between faith and skepticism into an empirical inquiry into environments, experience, mental simulation, and the limits of transfer.