Conditions for Intuitive Expertise: A Failure to Disagree

The question

When can an expert’s gut instinct actually be trusted?

The method

An adversarial collaboration [1] — not new data, but a joint effort by two researchers who’d spent careers disagreeing (Kahneman, skeptical of intuition; Klein, a champion of it) [1] to find where decades of judgment research actually converge.

The findings

Skilled intuition is recognition, in Herbert Simon’s phrase quoted approvingly by both authors [1] — “the situation has provided a cue… intuition is nothing more and nothing less than recognition.” Genuine intuitive skill develops only under two conditions: the environment is regular enough to contain valid, learnable cues (a “high-validity environment”) [1], and the person has had enough practice with fast, clear feedback to learn those cues [1]. Where either condition is missing — long-range political forecasting, picking individual stocks [1] — experience produces confident judgment without accurate judgment, as in Tetlock’s forecasting research [1]. The authors’ sharpest warning: subjective confidence does not distinguish skilled intuition from the heuristic-driven kind — “there is no subjective marker that distinguishes correct intuitions from intuitions that are produced by highly imperfect heuristics” [1].

The limits

This is a synthesis and position paper, not a new experiment — its evidence base is decades of other researchers’ studies (chess masters [2], firefighters [1], NICU nurses [1], loan officers [3], weather forecasters [4]) that the authors reinterpret through the validity-plus-feedback framework. The two-condition test is a well-supported heuristic for judging when to trust intuition, not itself a freshly tested finding.

• When an experienced hiring manager says “trust me, I can just tell” about a candidate from an unstructured conversation (see Knowing what great looks like): ask whether the interview gives them valid cues and whether they get fast, clear feedback on past calls. Most interviewers never learn whether their first impression was right, because nothing tells them — which is close to the zero-feedback case the paper says produces false confidence, not skill.
• When a functional expert makes a fast call inside their own domain — an engineer sizing up a system, a salesperson reading a deal: their gut is worth taking seriously. Deep, repeated, fast-feedback experience is exactly the condition the paper says produces real skilled intuition.
• When someone delivers a confident multi-year strategic forecast on gut feel: this is closer to the paper’s zero-validity case — long-range prediction with no fast feedback loop — where confident intuition is the kind Kahneman and Klein warn is least trustworthy, however senior the person delivering it.