Conservative Bayesian or natural stupidity?

It is a lot of fun to read Michael Lewis's new book on the collaboration and friendship between Danniel Kahneman and Amos Tversky.

Kahneman is portrayed as an "outsider." Before he met Tversky, according to former colleagues and students, "he was very insecure," "he was moody in the extreme," and "he was like Woody Allen, without the humor." His deep-rooted skepticism even applies to himself. After completing the draft of his now best-seller "Thinking, fast and slow," he hired someone to read his work to convince him not to publish it.

Tversky, on the other hand, was an "insider." One of his long-time friends described, "people who knew Amos could talk of nothing else. There was nothing we liked to do more than to get together and talk about him, over and over and over." He was the center of attention in every party. People loved to be around him and listen to what he had to say.

Yet, these two strikingly different characters and similarly remarkable minds met and formed a friendship that has fundamentally transformed how we view our own minds. After Tversky gave his talk on the "conservative Bayesians" in Kahneman's seminar, Kahneman immediately casted doubt on this notion that we are naturally Bayesian thinkers who constantly update our own beliefs with new pieces of information coming in. Kahneman's reaction to Tversky's talk, "Brilliant talk, but I don't believe a word of it," really unsettled this otherwise confident, smartest person in any room. The two then set off on an intellectual journey to study "natural stupidity."

Now, we know our minds are subject to many biases and heuristics, or mental shortcuts. We hate losses more than we like gains.