4 ms·

The guy who Tversky and Kahneman really hated, Gerd Gigerenzer had a very good answer to a lot of their work. He pointed out that: A critic of the work of Dani

by sien 8y ago

The guy who Tversky and Kahneman really hated, Gerd Gigerenzer had a very good answer to a lot of their work. He pointed out that: A critic of the work of Daniel Kahneman and Amos Tversky, Gigerenzer argues that heuristics should not lead us to conceive of human thinking as riddled with irrational cognitive biases, but rather to conceive rationality as an adaptive tool that is not identical to the rules of formal logic or the probability calculus. He and his collaborators have theoretically and experimentally shown that many so-called cognitive fallacies are better understood as adaptive responses to a world of uncertainty—such as the conjunction fallacy, the base rate fallacy, and overconfidence.[4] https://en.wikipedia.org/wiki/Gerd_Gigerenzer https://en.wikipedia.org/wiki/Gerd_Gigerenzer His books are well worth a read. Nassim Taleb also has a similar criticism in that he says that Kahneman and Tversky essentially said that humans don't act according to theoretical rules but instead have their own heuristics that have been derived from dealing with an uncertain world throughout history: https://twitter.com/nntaleb/status/883689826335281153?lang=en https://twitter.com/nntaleb/status/883689826335281153?lang=e...

Balgair 8y ago
Rationality is somewhat of a new concept to humans. It's somewhat surprising that economists took it so literally for that long and put and any real faith in it. The lemma of rationality does make the math a LOT easier, and we're doing better now about the irrationality aspects, but that spurt of faith in rationalism was quite interesting. For the large majority of our history, we've not been rational in the least. HN had a good discussion [0] on the Medieval Mindset [1] earlier in the summer. For ~1000 years in the Medieval period, the main mindsets were not rationality vs. irrationality, right vs. less-wrong vs. wrong, etc. But more Pious vs Impious, Cruel vs Kind, The Ideal vs the Real, etc. The people were no less people, but their heads weren't ours. What economic theories will come next, what new ways of thinking, what new mindsets? We're so focused on the rational, the 'right' answers, these days. But life, as we all know, is VERY stochastic ( a fancy word for random ). Maybe new mindsets about the randomness, bounded and given standards of deviation, will be a new paradigm, not just in economics, but in everything we do. [0] https://news.ycombinator.com/item?id=17058487 https://news.ycombinator.com/item?id=17058487 [1] https://coinsandscrolls.blogspot.com/2017/09/thinking-medieval-seeking-endarkenment.html https://coinsandscrolls.blogspot.com/2017/09/thinking-mediev...
pongogogo 8y ago
I parsed Taleb differently to you, I think part of his point is most behavioural economics is only looking at whether behaviour is rational or not and the individual level, but they are missing the fact that we're a herd species and behaviour has meaning as well in aggregate - there's second order effects. His point about nudges etc is that it is dangerous, as we focus on first order effects but ignore second order effects. This is what creates large risks, for example, defaulting everyone into one type of pension plan which is itself defaulted into one style of index investing, most of the indexes use the same tracking algorithm creates second order risks that we don't understand where there were none previously.

by sien 8y ago

Balgair 8y ago

Rationality is somewhat of a new concept to humans. It's somewhat surprising that economists took it so literally for that long and put and any real faith in it. The lemma of rationality does make the math a LOT easier, and we're doing better now about the irrationality aspects, but that spurt of faith in rationalism was quite interesting. For the large majority of our history, we've not been rational in the least. HN had a good discussion [0] on the Medieval Mindset [1] earlier in the summer. For ~1000 years in the Medieval period, the main mindsets were not rationality vs. irrationality, right vs. less-wrong vs. wrong, etc. But more Pious vs Impious, Cruel vs Kind, The Ideal vs the Real, etc. The people were no less people, but their heads weren't ours. What economic theories will come next, what new ways of thinking, what new mindsets? We're so focused on the rational, the 'right' answers, these days. But life, as we all know, is VERY stochastic ( a fancy word for random ). Maybe new mindsets about the randomness, bounded and given standards of deviation, will be a new paradigm, not just in economics, but in everything we do. [0] https://news.ycombinator.com/item?id=17058487 https://news.ycombinator.com/item?id=17058487 [1] https://coinsandscrolls.blogspot.com/2017/09/thinking-medieval-seeking-endarkenment.html https://coinsandscrolls.blogspot.com/2017/09/thinking-mediev...

pongogogo 8y ago

I parsed Taleb differently to you, I think part of his point is most behavioural economics is only looking at whether behaviour is rational or not and the individual level, but they are missing the fact that we're a herd species and behaviour has meaning as well in aggregate - there's second order effects. His point about nudges etc is that it is dangerous, as we focus on first order effects but ignore second order effects. This is what creates large risks, for example, defaulting everyone into one type of pension plan which is itself defaulted into one style of index investing, most of the indexes use the same tracking algorithm creates second order risks that we don't understand where there were none previously.