Thinking, Fast and Slow — Ch 22: Expert Intuition: When Can We Trust It?

Expert Intuition: When Can We Trust It?

Chapter 22 is the book’s crown jewel. The Kahneman-Klein framework for when to trust expert intuition – high-validity environment plus opportunity to learn regularities – is widely accepted and practically useful. Meehl’s finding that algorithms beat experts in structured tasks has been confirmed across 136+ studies. Tetlock’s forecasting research shows why political pundits fail: low-validity environment with ambiguous feedback.

So what

This chapter presents the book’s most practically useful framework: when to trust expert intuition and when not to. The Kahneman-Klein conditions are widely accepted and provide a diagnostic tool applicable to medical training, military decision-making, and hiring. If you read only one chapter, make it this one.

Verdict

Clean

Claims checked in this chapter (3)

holds✓ verified

Expert intuition is valid only when two conditions are met: (1) a high-validity environment with stable regularities, and (2) adequate opportunity to learn those regularities through practice with feedback.

holds✓ verified

Simple statistical models consistently outperform expert clinical judgment in structured prediction tasks, as demonstrated by Meehl (1954) and confirmed by subsequent meta-analyses.

holds✓ verified

Expert political and economic forecasters perform barely better than chance, and pundits with the most media visibility tend to be the least accurate (Tetlock, 2005).