Richard's Reviews > Thinking, Fast and Slow

Kahneman is a brilliant thinker who's small experiments cracked the foundations of neo-classical/rational choice theory. His collections like Choices, Values and Frames are terrific reading, full of extraordinarily disorienting observations of actual human behavior.

That said, this book was a bit of a disappointment. In this book, Kahneman weaves together a series of ideas based on his experiments and the experiments of like-minded researchers, and the effect is underwhelming. Gone is the careful exploration of small twists in human behavior, replaced with broad, rolling declarations that are only weakly supported by the data. For most of the experiments he cites, he fails to talk about the power of the findings (were participants 40% more likely to do something when framed a certain way? Or 4% more likely? Or 400%?). Sure, this is a book for a wider audience than his journal articles or academic collections, but I found that the dumbing down went too far.

Too bad. The basic research that made Kahneman a star is fabulous. Loss Aversion in Riskless Choice is a great example - a review of experiments involving no more than some college students and coffee mugs, with counterintuitive findings about loss and market prices. Kahneman is at his best when he lets the reader figure out the implications of his stunning findings.

Also, his dual process theory of thought - System 1 and System 2 - seems the less powerful of two competing views of human reason. A skeptic might ask what evidence is there that System 2 exists at all. System 2 certainly seems to be no less vulnerable to confirmation bias than System 1. The Mercier and Sperber formulation on the function of human reason just fits better with evolutionary models. Of course, that's probably just my confirmation bias kicking in.