In the laboratory, people overweight rare events — buying lottery tickets, purchasing insurance against improbable disasters. In real life, people underweight rare events — ignoring small risks until they happen. The contradiction has a resolution: how you encounter the probability determines how you weight it.

The Framework

Choice from description means deciding based on stated probabilities ('there's a 5% chance of X'). Choice from experience means deciding based on personal sampling of outcomes (trying a restaurant 20 times and getting food poisoning once). The two produce opposite biases: described probabilities are overweighted (possibility effect), while experienced probabilities are underweighted (because rare events may never appear in a limited personal sample).

The resolution: in description-based choice, the stated probability triggers the possibility/certainty effects that prospect theory documents. In experience-based choice, the probability must be encountered — and in a small personal sample, a 5% event may never occur. If it hasn't occurred in your experience, it effectively has a decision weight of zero. This explains why people continue to build in flood plains (they've never personally experienced a flood) despite knowing the statistical risk.

Where It Comes From

Chapter 30 of Thinking, Fast and Slow presents the description-experience gap alongside Ralph Hertwig's research on experienced probabilities. The finding undermines the universal applicability of prospect theory's probability weighting: the theory describes decisions from description accurately but less accurately describes decisions from experience, where rare events are underweighted rather than overweighted.

> "From experience, people tend to underweight the probabilities of rare events." — Thinking, Fast and Slow, Ch 30

The Implementation Playbook

Risk Communication: If your audience will make description-based choices (reading a brochure, evaluating a proposal), vivid presentation of rare risks will be overweighted. If they'll make experience-based choices (ongoing product use, daily commuting), rare risks will be underweighted until personally experienced.

Product Safety: Users who have never experienced a security breach underweight cyber risk (experience-based). Users who read about breaches overweight it (description-based). Security messaging should bridge the gap: make the described risk vivid enough to overcome experience-based underweighting.

Insurance Sales: Homeowners who have never experienced a flood underweight flood risk. Showing them flood statistics (description) may overweight it. The most effective approach: make the description vivid enough to overcome experiential underweighting without triggering excessive fear.

Key Takeaway

The description-experience gap means that prospect theory's probability weighting is context-dependent. Described probabilities are overweighted; experienced probabilities are underweighted. Your communication strategy should match the audience's decision mode.

Continue Exploring

[[Possibility Effect]] — The overweighting of small described probabilities

[[Availability Heuristic]] — The mechanism by which experienced events become available

[[Denominator Neglect]] — How the format of described probabilities amplifies the overweighting

📚 From Thinking, Fast and Slow by Daniel Kahneman — Get the book