Hi Everyone,

I was discussing LessWrong and rationality with a few people the other day, and I hit upon a common snag in the conversation.

My conversation partners agreed that rationality is a good idea in general, agreed that there are things you personally can do to improve your decision-making. But their point of view was that, while this is a nice ideal to strive to for yourself, there's little progress that could be made in the general population, who will remain irrational. Since one of the missions of CFAR/LW is to raise the sanity waterline, this is of course a problem.

So here's my question, something I was unable to think of in the spur of the argument - what are good examples of rationality techniques that have already become commonly used in the general population? E.g., one could say "the scientific method", which is certainly a kind of rationality technique that's going semi-wide adoption (though nowhere near universal). Are there any other examples? If you send a random from today back in time, other than specific advances in science, will there be anything they could teach people from the old days in terms of general thinking?

New to LessWrong?

New Comment
31 comments, sorted by Click to highlight new comments since: Today at 4:10 AM

Small rationality-inducing cognitive algorithms that manifest in everyday speech:

  1. " You're missing the forest for the trees." Given what you want to achieve, spending time on these details may not matter.
  2. "For example?"
  3. "What's the chance that that'll actually happen?" Think about the probability of the hypothetical. The notion that randomness and chance is something that can be measured is a rather recent phenomenon in human history.
  4. "What's done is done." or "No use crying over spilt milk." Don't fall prey to the sunk-cost fallacy.
  5. "I'm going to play the devil's advocate here." Steelmanning your argument.
  6. "What difference does it make?" Does what we're arguing about actually have any consequences?
  7. "You've got to take the long-term view." Don't think of the action you're considering as a single-shot thing, but more of as part of a strategy.
  8. "They did this study..." What does the literature say? I agree that this move is often used to just support the argument you want supported; but hey, it's a step in the right direction.
  9. "I don't get it." I notice that I'm confused.
  10. "So let's be clear here." OR "Just to make sure that we're on the same page here." You and I may have different models, so let's ensure that we have the same model.
  11. "This is win-win." Basically, a Pareto improvement.
  12. "You do your job. I'll do mine." Benefits of specialization.

I agree that these phrases aren't very nuanced or advanced rationality techniques, but they're steps in the right direction. And to better is why we have things like Less Wrong.

Also, "Care to bet on it?" OR "Why don't you put your money where your mouth is?" OR "I'm willing to bet that..."

Making lists.

A/B testing and data-driven corporate decision-making in general. Cost-benefit analysis. Discounted cash flow analysis. Arithmetic. Expected value (it tells you not to play slot machines).

[-]chrt10y60

Expected value (it tells you not to play slot machines)

Casinos are apparently still making money, so I question the extent to which this has been adopted by the Masses.

That just means that the sanity waterline isn't high enough that casinos have no customers -- it could be the case that there used to be lots of people who went to casinos, and the waterline has been rising, and now there are fewer people who do.

Deciding to play slot machines is not a choice people make because they think it will net them money, it's a choice they make because they think it will be fun.

I do not think that this is true for the majority of players.

Sure. Probably lotteries are a better example. My guess is that most people edanm talks to about this topic will consider themselves too intelligent to play the lottery.

Perhaps, but I'm trying to convince intelligent people that there are real changes we can introduce that will be adopted by most people, so I'm not sure the lottery fits the bill.

I'm trying to convince intelligent people that there are real changes we can introduce that will be adopted by most people

Well you can start with trying to convince me then. If we can't get people to stop playing the lottery, what makes you think we can get them to understand and correctly apply Bayes' Rule? BTW, I hope you don't have the bottom line already written here.

I believe there are meaningful things people believe/do nowadays that they didn't 300 years ago (e.g. using the scientific method).

Unfortunately, for all these things, they're either: a) adopted only by some people, not the majority. b) As DanArmak says, adopted only because of "peer pressure" or other social reasons.

Now, that's not to say CFAR's mission isn't still worthwhile - raising the sanity waterline of just certain segments of the population, e.g. the top X% in terms of intelligence, is still of great importance.

But if there really aren't general "rationality techniques" that have been adopted by most people, if the average person today is no more rational than a person 500 years ago, then I suppose you're right - my bottom line might need to change to "maybe we can't reach the general populace".

what are good examples of rationality techniques that have already become commonly used in the general population?

Language. Arithmetic. Learning stuff. Modelling other people's minds.

Rationality techniques commonly used will not look like anything special, simply because they are common. Unless one looks through the lens of this exercise.

Does the internet count as "the general population"? If so: identifying and shaming logical fallacies. Sure, people do it imperfectly, and a lot more readily for the opposing side than for themselves, arguments are soldiers etc. But it's still harder to get away with them, for an overall positive result on truth-seeking.

One example that many people have adopted is googling things. It used to be that if people disagreed about some random fact (like if Norway or Iceland had more people) they would argue about it. Now someone will google it pretty quickly. This is widespread enough that it might not even feel like a rationality technique, but it is, and a powerful one at that.

This shows that people will adopt them, and do it in big enough numbers that it seems normal as long as it is made sufficiently easy.

"Making a list of Pros and Cons" is a very basic step toward rational decision making. Implicit in the method is the idea of assigning weight or utility to each thing on the list, since the idea is to "weigh the outcomes and make a decision". Throw in probability assignments and quantify the utility estimates instead of doing the weighing in your head and you get an expected utility calculation.

Implicit in the method is the idea of assigning weight or utility to each thing on the list, since the idea is to "weigh the outcomes and make a decision". Throw in probability assignments and quantify the utility estimates instead of doing the weighing in your head and you get an expected utility calculation.

That, I would say, is a very narrow idea of rational decision-making. Quite impractical one, too.

That, I would say, is a very narrow idea of rational decision-making.

Well, yeah. I did say it was a very basic step, not a fully-vetted rationality method. But making a list of pros and cons is better than making decisions based on emotions or impulse and it's common enough that almost everyone is familiar with it. And it's a place to start teaching people how to do it better. The only other non-corporate examples I see here are "making lists" and "nigerian scammers" which are also very basic. I feel like it answered the request which was for "rationality techniques that have already become commonly used in the general population".

If the general population already had a lot of really good rationality techniques in common use, this wouldn't be an interesting discussion.

I think we should make an important distinction here: why do people adopt en masse new beliefs and behaviors, rational or not?

Do they adopt them because they become convinced of their truth, or by considering the evidence for and against them? Or do they adopt them for the same reasons they adopted the irrational ideas they are replacing - through cultural effects, following prestigious leaders and popular opinion?

I am not, by the way, trying to denigrate consensus-following. It's quite a good heuristic. But it's not what we usually mean by "rational behavior"; it is more nearly instinctive than conscious.

That's a very good point, although I think a good a first stage is to find what techniques people are actually using, then try and understand why.

A good second stage is to look for techniques that were publicized and not used, and see why some techniques gained currency while others did not.

A lot of decision making techniques that people are actually using aren't "rationality techniques". Focusing on "rationality techniques" means that you don't count that churches get their members to pray.

If the churches have good strategies for getting their members to engage into practicing certain behavior copying those techniques might be better.

I see what you're getting at, although praying is a bad example - most people pray because their parents and community prayed, and we're looking at ways to lead people away from what their parents and community had done. The Protestant Reformation might be a better case study, or the rise of Biblical literalism, or the abandonment of the prohibition on Christians lending money at interest.

Inquiring after the falsifiability of a theory?

Not perfect but very good, and pretty popular.

Any claim made in an advertisement is automatically suspect.

Paper. Calendars. Smartphones. Asking other people for advice.

Isn't that just technological progress? Except for asking people for advice, nothing else there changes how people think, so it's hard to call it a rationality technique IMO.

I totally disagree! Paper is an amazing tool for changing how people think: it gives them an external source of working memory!

[-]Gav10y10

Perhaps Nigerian scammers? Nowadays everyone knows that Nigerian scammers are likely going to take you for a ride, and what tactics they use. And how much it hurts to get stung by one.

You could use the idea of benevolent Nigerian Princes as the 'known false' idea that everyone can agree on, and segue into the features of it that ring alarm bells (high payoff for sending personal info, undeserved wealth triggering our 'greed' buttons, etc) and the idea that being primed on rationality helps you to not fall for whatever the next version of Nigerian scamming turns out to be.

[-][anonymous]9y00

Every time a parent has to eat something a child left on the plate, saying s/he has 'eaten everything', the concept of actively seeking disconfirming evidence taps on the door...

Epistemic or instrumental rationality?