185. Bounded Rationality | THUNK

Many academic fields assume that maximizing utility is synonymous with rationality (& marvel at how *bad* humans are at it). But the real world very rarely allows such clean analysis - maybe we're not so irrational after all?

- Links for the Curious-

Irrational time allocation in decision-making | Proceedings of the Royal Society B: Biological Sciences -

[catalyst]

When I was a younger fellow I suffered from chronic analysis- paralysis. I was a perfectionist even when it came to mundane tasks. It gave me great anxiety because I felt like no matter I did I there might have been some way to do it better. So i wasted a lot of time unable to decide what to do, to the point I couldn't complete much of anything. Nowadays, I go with the flow, but still maintain an idea of what direction to go.

[AP-yxmm]

Great Video!I was waiting for something like this topic ever since the streaming session, thank you! I remember writing you about Gigerenzer, who is interested more specifically in the heuristics. Another interesting topic is framing, like the research that George Lakoff has done. Further step from this point is also the issue of ethics of such a way of thinking, if and how can it be ethically challenging. Really good job with this video!Stay healthy!

[YuriRadavchuk]

There is a whole universe of modified rationality concepts in the philosophy of science. Thank you for bringing this up!

[passingthetorch]

This is an extremely important concept. It is important to know that we have cognitive biases ... but it's just as important to ask why, and making "good enough" decisions under constraints of time and energy is probably the reason. We have so many tools, tricks, models, etc., to help us think either more clearly, more precisely or more quickly (language, mathematics, cognitive biases, etc.) But there is the famous saying: "All models are wrong, but some of them are useful." I would suggest that the OODA loop is something we have more or less evolved to optimize. Finally, bounded rationality should teach us to be humble ... I can barely optimize my choices for long term happiness ... why would I try to optimize for large numbers of people? Simply giving them the freedom to optimize their own lives (subject to the restriction of not actively harming others), is a much humbler proposition.

[KlaustoFausto]

I love this channel to pieces. Nothing more to say

[bubbles]

Really interesting episode! :D kinda reminds me of the book 'The Paradox of Choice' by Barry Schwartz, about how modern society is plagued by so many choices that it not only results in people being stuck in analysis paralysis but also when it comes to making decisions produces an enormous amount of stress, there's a pretty good TED talk about it by the same author too

[PetersonSilva]

Thanks for the video! Awesome as always!

[nicholashughes]

Really great episode really well presented Thanks!

[davidpetry]

A really interesting video. If I had more time to think about it I probably have more to say. Instead I just blurted out the first thing that came to mind. Sorry

[catalyst]

Look before you leap but don't look down.

[KohuGaly]

Is there really anyone who interprets utility maximizing so naively? I mean... the moment I saw it, it was clear to me that one has to apply reasoning about utility recursively upon itself (a sort of meta-reasoning about utility), because the real world limitations are another factor to optimize.
In most cases, one only has to do the meta-reasoning once, to build a heuristics with easily identifiable conditions. And one can update the heuristic over time to sharpen it further, if needed. Instead of holding to your biases and heuristics mindlessly, it's a good practice to reexamine your past decisions, once you are no longer constrained by time or resources, to check if your heuristic in that situation is optimal and update it if need be.

[random]

My Fitness Pal suggests breakfast iirc. What should you have for breakfast is a question asked by each individual tens of thousands of times There are billions of individuals. There should be an app for this problem. It would not be analysis paralysis for years spent on such a problem if the utility to others (who pay) would be that great.

[tochoXK]

I've read "Thinking Fast and Slow" by Daniel Kahneman (who is the one who discovered "cognitive biases"), and he kinda agrees with these points.

[KlaustoFausto]

I guess the decision on whether to act 'fully' rational and when to act 'good enough' (using biases etc) should depend on the importance of the decision on top of the other boundaries like urgency and brainpower.
Using your gut feeling to vote is quite a bad idea, since tribalism and other biases will undoubtedly influence your decision, but when choosing which cake to eat it's just not worth it to devote your concentration and time to this task, and the damage done will be a lot smaller if you're choosing wrongly.

[landspide]

Yep, never underestimate intuition...

[clearmenser]

Anyone a fan of The Secretary Problem?

[judgeomega]

10:12 "cogitating" i see your thesaurus isnt on sick leave.