Thinking, Fast and Slow

Kahneman‘s 2011 book (Farrar, Straus and Giroux; research and writing help from Jason Zweig), the popular summation of the heuristics-and-biases programme and the book that put System 1 / System 2 into general circulation. Over a million copies by 2012; 2012 National Academies Communication Award.

Read via the Wikipedia article, not the book. That matters for what this page can claim: it records the book’s structure and its reception accurately, and none of its findings first-hand.

What’s in it

• Two systems. System 1 is fast, automatic and emotional; System 2 slow, deliberate and effortful. Most of the book is about System 2 ratifying System 1’s answer instead of checking it.
• Heuristics and biases — anchoring (an irrelevant number moves a judgment), availability (probability estimated by ease of recall), the conjunction fallacy (the Linda problem).
• WYSIATI — “what you see is all there is.” The mind builds a coherent story from available evidence and doesn’t represent what’s missing, which is the general engine under the focusing-illusion.
• Prospect theory — the Nobel work with Tversky: gains and losses valued asymmetrically, against a reference point, with losses looming larger.
• Two selves — the experiencing self who lives moment to moment and the remembering self who evaluates afterwards. Kahneman: “I am my remembering self, and the experiencing self, who does my living, is like a stranger to me.”
• Peak-end rule — the remembering self scores an episode by its peak and its ending, largely ignoring how long it lasted.

The two-selves split is the deep version of what kahneman-focusing-illusion reports second-hand: the income studies disagree because asking queries the remembering self while experience sampling queries the other one. They aren’t rival measurements of one quantity; they measure two different things that both have a claim to the word “happiness.”

The replication problem, in the book’s own bibliography

This is why the page matters more than a summary would. Analysis of the studies cited in Chapter 4 (“The Associative Machine”) put its replicability index at 14 — essentially no reliability. Broader analyses argue much of the book rests on literature with shaky foundations, with the social-priming work worst affected.

Kahneman’s response is the part worth keeping: “I placed too much faith in underpowered studies.” He conceded it publicly rather than defending the chapter.

The irony has been widely noted — a book about systematic error in judgment, whose author made a systematic error of judgment about which evidence to trust. Read generously, that’s the strongest possible illustration of its own thesis, and the reason replication-crisis now has its own page here. Read strictly, it means a claim’s presence in this book is not evidence for it, and this wiki should cite the underlying study or say it hasn’t.

Related

daniel-kahneman · heuristics-and-biases · focusing-illusion · replication-crisis · kahneman-focusing-illusion · synthesis