Two rationality tests

If you were trying to assess a person’s rationality on the basis of one not-directly-verbal indicator, given his or her behavior over the course of a meal, what would it be?

And if you could ask only one question of a person, to assess his or her rationality, which question would it be?

That’s from me!  As for the first benchmark, you cannot refer to verbal answers to questions you might ask.  You could however nominate “the person hesitated for a long time before answering each question,” or something similar along the behavioral dimension.  That is what I mean by “not-directly-verbal.”

Part of me wishes to suggest “are they carrying a book or not?”, but alas too many semi-rational people don’t do that.  I might consider the process by which they select a menu item and order their food, as a kind of proxy for decision-making more generally.  How well they treat the server would be another variable of interest.

As for the second question, I suggest asking the person who he or she thinks are the rational people.  If the answer is considered and uncertain and complex, upgrade the rationality of that person.  If the answer is dogmatic and refers to holding a particular doctrine…

I considered asking the person if he is himself rational, but that simply will induce lying and false modesty.

Can you think of better tests?


For the second, "Would you be happy if you received an income tax refund?"

No, I hate the thought of giving the government an interest-free loan and I abhor filing ASAP instead of waiting until 15 April.

Ah, but if you wait until 15 April, are you not simply giving them an interest-free loan for a longer period of time?

Take enough withholdings to owe the government money come tax time (but not enough to get hit with an underpayment penalty) and file April 15th.

In fact, I made a Form 1040ES payment before 18 April. If the Form 1040 amount due is more than 10% of total tax due, there are interest and penalties.

I rough out the year's taxes as soon as I receive all the 1099's, etc. I surprised myself when I found the IRS owed me.

I received the refund in early March.

The typical plan is to owe some to the IRS. We have some income subject to withholding and some not. I over-estimated the 2016 1040 ES payments. So far in 2017, I am remitting lower 1040ES quarterly remits.

I'm with you, anything that starves the beast is preferable. Another example is the US Post Office. I would do anything to not pay one penny more than the postage rate for first class mail. I hated the thought of the government getting any more of my money.

I understand what you're saying, but that question suffers from two problems.

1. Some people pay the correct amount in taxes given their situation in the time, and then their situation changes unpredictably (e.g. getting fired) over the year, leading to a tax refund. In that case, it would be rational to be happy to get a tax refund.

2. Given that I have paid too much in taxes, of course I would be happy to get a refund. I'd be happier if I hadn't overpaid in the first place, but I much prefer overpaying and getting a refund to overpaying and not getting a refund.

It also can come off as "Stop having fun a way I don't like!"

If I hide a dollar in a jar every week and on New Year's I open the jar and am happy with my $52, what's the problem? It's a much cheaper form of entertainment than most other things. Plus it lets me save money. Everything here is a winner.

Each time I squander money (and pay cash) on food, clothing, etc., I put the change in the Lotto Reserve Fund. Some day, it will be beer and pretzels for everybody!

Alternatively, "Yes, I'm poor, and will receive more in end-of-year taxes than I paid in the first place."

How about, within a certain band (+/- $500 - $10k depending on marginal value of a saved dollar) I don't care if I get a refund or not. Precisely optimizing my withholding through income, deduction changes etc holds little value for me and the opportunity cost is too great. I'd rather read another few entries on my favorite economics blog.

I agree. In this extremely low interest rate environment, I find it far more cost effective to overpay estimates, extensions, etc., and true up with the final return (I always file in October), and even that, I typically just roll into the next year. If you look at how much that loan "costs" you, your time is worth a lot more. (I have an accountant, but still, to not overpay, it's like doing your taxes well over 2.5 or more times each year (once with the extension, once with the actual filing, and the estimates are hard because for the first ones my prior year return is not done so I don't know what 110% of that is). Overpaying saves a lot of time. (Plus if you try to cut it close (not overpay) and miss, the penalties also cost you a lot more than the loan cost you.)

We need someone intelligent to weigh in on this, is Barkley Rosser available to give his two cents?

Barkley isn't even top quartile around here. Or was that the joke?

One, how frequently the person used the napkin.

Two, "Did you vote for Hillary?"

In which Dick proves he is not rational by wasting his question trying to assess if the person holds a particular doctrine.

And The Other Jim proves he is not rational by wasting his time replying to an obvious troll post.

I see an infinite regress here ;-)


The second question IS redundant. The tear stains always tell. Anyhow, some people never figure out it's time to leave the funeral.

Time would be a factor in #1. Putting what he or she said aside, simply observing someone closely over the course of a long dinner at a public place is telling. So, my answer is not to look for a specific indicator of rationality but for the absence of indicators of irrationality during the meal. The longer the time elapses before something that strikes you as irrational, the greater the probable level of overall rationality.

For #2, since I consider it a difficult question to find a good answer for, especially if it is someone I know not at all, I would just ask them the same question and see what I thought of their answer. The problem with this approach is that a crazy person might very well be able to give a terrific and impressive answer.

Ask them

"What do you think I should order?"

Then analyse their approach ...

Tricky.... People are rational about some things and not rational about others. The non-rational subjects usually involve emotional reactions.

1) Whether or not they compulsively checked their smartphone
2) Something to test sunk cost fallacy (one of most common forms of irrationality) - "I know we ordered white wine, but now I'm in the mood for red, should we order another bottle?" or something of the sort

out of curiosity, what do you think the answer is to your red-white question? I say the white is free (already paid for) whereas a bottle of red will cost money - so I'll drink the white even if I enjoy it slightly less than the red, then decide afterwards whether to purchase another bottle.

I have a sneaky suspicion that the sunk-cost fallacy is among the most over-cited and least-analyzed economic concepts in existence.

Yes, exactly. Unless the sudden preference for red is so overwhelming that a bottle of wine is the same as having no wine at all, this example wouldn't be a case of the sunk cost fallacy. And I have also gotten the impression that most invocations of the fallacy aren't actually describing it.

You're right Clay: the red-white question is not likely to be an example of the sunk cost fallacy. Assuming the white wine has actually been purchased and cannot be returned for a refund, the fallacy would apply if the person felt compelled to drink the wine even if he didn't want it any more, meaning that drinking it would displease him. Alternatively, it would apply if the person eschewed the red even though the marginal utility to him of drinking the red (red utility minus white utility) exceeded the price of the red. Merely preferring the red to the white is not enough; one must compare the marginal improvement in utility from switching wines to the marginal cost of doing so.

I would have said "red" but I see your point. It all depends on how much value you place on the white. If you weren't going to drink it at all, then it would be irrational not to get the red, but if it would still be acceptable, it's less clear.

I would think: For the love of God, by the time the red gets here, you'll be in the mood for something else. Stop being a slave to your mercurial moods. But sure, order the red.

All of the above would be subject to:

Are you paying? Get whatever you want; knock yourself out.

Am I paying? Leads us to: How much do I like you? Do I still think your whimsy is cute? Sure, get the red. Has your inability to confidently make a decision and stick to it finally worn out its welcome? Hell, no, we're not ordering the red.

Are we splitting? Do *I* want the red? [Hypothetically, of course, since I don't drink and I'm already subsidizing your imbibing. But fine, I'll just make sure that the appetizer I prefer comes out on top.] Now we have to dig deeper: how much do I care about this relationship? Am I here by choice, or is this a work or social obligation? Do I need something from you? How much does the wine cost? Am I enjoying your company?

Is it being expensed? Well. I do know now that you are very profligate with other people's resources to satisfy your own petty desires. That tells me all I need to know.

Too many factors. I guess I'd just be the person who sits there quietly for awhile,

Being overly focused on the color / flavor of the wine, given that it is within an acceptable range, while having an otherwise interesting dinner companion would lower one's rationality score for me.

I'd ask nothing, just listen.

But, I'd select a high traffic coffee shop. In that setting there are lots of stimuli that makes the person reveal their mind. Some people will start to criticize the way of dressing or talking of others around. Any kind of expectation not met is very revealing: people will express dissatisfaction because WiFi is slow, the place is loud, chairs not comfortable, bad coffee quality, etc.

The rationality test is just the ability to (mostly) ignore all the stimuli around and focus on the conversation we're having. It doesn't matter if we're just talking about the weather.

This sounds like more of a test of whininess than rationality. Tyler Cowen might even think the non-whiny diner is complacent, which he regards as a sort of short-sighted irrationality.

Yes, abstract thinking can be described as non-whiny way of thinking. Ideas more important than events.

I complain about how loud places are, but I completely understand in certain contexts I am the one that is the problem. Coffee shops should be buzzing and alive with energy (not sure why, some norms/status quo bias involved there). A loud library - as an extreme example - is not irrational to complain about. At least to me...

Rationality is becoming a religious affiliation, so look for unusual rituals.

Ask to see the tattoo.

And start by mistrusting yourself. Anyone who would carry on social interactions at this level is irrational to start with, so how could an irrational person be able to judge 'rationality' in someone else?

If they agonize over their food choice they are probably irrational. Food isn't very important.

This is a fantastic takedown, congratulations.

Agreed - and it also shows they have not previously put in the legwork to form their priors.

But the things that you agonize over are very important, of course, right?

1. Most measures of behavior have a low R value, so might as well go for the stereotyped indicator of flat affect.
2. Assuming people are not equally rational about all topics, I'd ask an oft discussed question on the topic I care about, listening for the cues you discuss above.

When the waiter asks if anyone wants dessert, if X says no, I would then say "yes, I would like dessert".
If X changes their mind and then asks for dessert, they're probably not completely rational.

Pretend to receive a call on your phone, and then offer the phone to the other person and say "It's your (Dad|Mom|Wife|Husband|Son|Daughter)". If they respond with "my Dad's not a phone", they're probably not very rational

'they’re probably not completely rational'

Or merely polite.

No, this is Patrick.

posting to raise the above commenter in status

Disagree. The choice of having dessert or not is more usually a function of time than of whether to have yet another creme brule.

The game of ordering dessert or not, understood as a signal of whether the meal, and conversation, should be extended is complex.

'but alas too many semi-rational people don’t do that'

As always, this web site provides the sort of subtle amusement rarely found otherwise - just trying to grasp the idea that bringing a book to a meal with someone is a sign of something other than an action ranging from fairly explicit impoliteness to extreme rudeness is enough to jolt one's complacency at the idea of well mannered Ivy League alumni.

'How well they treat the server would be another variable of interest.'

Really, such subtle humor - how one is treated as a person sharing a meal with someone interested in judging their rationality is unimportant, but a well mannered approach to dealing with the paid staff is obviously a part of rationality. Or self-interest, of course.

'asking the person if he is himself rational, but that simply will induce lying and false modesty'

You really need to go out eating with a broader group of people. Starting with women, though apparently a pretty one as a meal companion would be a sign you are not eating in a place with food worthy of a leading light of the GMU econ dept. But this would also definitely include the sort of people that write books or perform in front of audiences, without at any point caring about whether they are 'rational' or not. Most artists do not value rationality, which makes sense, as their audience does not care much about rationality during a performance either.

'Can you think of better tests?'

Everyone is free to waste their time as they wish, of course, but this is just getting lazy.

Bringing a book to a meal means the person had a book to read along the way, and not that they plan to read during the email.

If someone had a book, I'd wonder why they don't just have their 9,000 favorite books available on their smartphone at all times. Carrying a physical book is more an indicator of aesthetic choices than whether they read or not. I'd put it in the same category of, do they order a cup of coffee, espresso, or a single-origin shade-grown pourover.

Nothing quite says "I love you" like a hot, wet tongue inserted deep up your farthole.

When I was young there were various "bitch-tests". Seems irrational to test the rationality of a date when a lack of bitchiness is more important.

Who wants to have dinner with a rational person? I don't. That would put me off the food as well.

Prof. Cowen is unlikely to be inviting you to a NoVa strip mall food experience any time soon. Loyal readers can decide on the rationality involved.

In fact, Dan and I have had a bunch of meals together.

"Does Dan Wang want to dine with you?" There's your test for rationality.

Dan Wang tickles my sphincter with his tongue. (It's hot and wet, I think he loves me.)

Congratulations - so which mood affiliation is the rational one when selecting the proper Straussian reading required for his comment?

(And congratulations again - for someone who only occasionally reads the comments, it is nice to see you have the chance to correct me instead of vice versa. You did catch the Post article on the Bose headphone app privacy lawsuit, compared to 2 year old news concerning eavesdropping TVs, right?)

Plot twist - they both had the taco bowl

Re-enact the opening scene from Reservoir Dogs: tell him you'll pick up the check if he can cover the tip. Watch what he orders and how he tips.

Similarly, and better in a group: state that the check will be split evenly regardless of who orders what. Does this increase or decrease the total check size?

1. I would take either extreme interest or disinterest in the food as potentially linked to high rationality.

2. I would ask them to describe the views of their ideological adversaries.

2: Do you think that people who vote 3rd-party for US President are wasting their vote?

Excellent! Though I fear that the number of people even with high IQs (admittedly not quite the same as rationality) who think that their votes "matter" in determining outcome elections is extremely high.

When people say voting matters, they are obviously making a moral claim.

Do you think politicians don't notice that they can win some votes when observing that several million people voted for a party or person with 0% probability of winning?

Those are dedicated voters. Whose votes may then be had. If they can be won.

If your vote doesn't really matter enormously, then most certainly one should vote for more moral than strategic reasons.

Excellent, because there is not one answer for every possible election. An explanation pro or con should show the gears turning.

No, there is one answer for every election. One vote never decides a federal election. So no vote is ever "wasted."

"People" obviously decide elections. Is it more than a cute trick to define an individual out of the "people" group?

2: How often are you aware of your cognitive biases?

I have none, but I am aware of evwryone else's cognitive bias.

Now this is the Thiago we have come to love.


Because you're a salad-tosser.

No, I am not.

"What is the least rational thing anyone said to you in the past week, and how did you respond?" Whether an answer follows, and what form it takes, might reveal a bit about the respondant's own rationality. Or, "Would you rather have $33,000 now or a dollar a year for a million years?"

It strikes me that "rationality" here is standing as a place holder for a whole package of complex behaviors, traits, and affiliations, some of which embed class and possibly race and gender (though maybe not).

Is what you are asking maybe closer to "does this person share Tyler's intellectual approach?" I'm genuinely not fully clear.

If so many other questions could get you more ground quicker.

Agree. Judgments of dining habits are going to be heavily culturally biased.

If they eat with their left hand they're disgusting.

Agreed. Use a fork!

As long as they eat my asshole out, that's all I care about.

I struggle a bit with defining rationality.

I guess we'd define it as someone who does Kahneman system 2 kind of analysis on *every question* in life. Because they can't do any snap judgement/system 1. Since system 1 pattern matching/aligning with the crowd is technically irrational. So perhaps if they sit in a chair until they starve to death before ordering dinner. Then they win.

More seriously, it seems you have to pick and choose which areas to focus on for rationality/system 2 analysis. So given this, perhaps the correct question is to ask which areas the person puts sustained effort into analyzing. Maybe it's their job, or a hobby, or Game of Thrones, or politics, or NBA basketball or unfriendly AI. But once they tell you their area of interest, then in that area they should have a nuanced view based on reasoning and reading about their topic of interest. Those views in some ways should not just disagree with their out-group, but also has some disagreement with their in-group. As in-group disagreement is the hardest, hence is the one to look for. If you are part of a community, but hold the same views as your community across the board, it's not clear that's a "rationalist" position, even if you are fortunate in your in-group (so you in fact hold many correct positions). We're talking rationality as a process to find truth, not as a scorecard for your in-group.

Here's an example. Socialist Freddie deBoer defending heritability of IQ. He gives the position of the American Psychological Association pretty much (as laid out on wikipedia). But that's not the position of some people in his in-group. So he was attacked him mercilessly.

So to me disagreeing with your in-group, not because you are ornery and like to troll, but because you've read up on a subject, and think you have a more accurate/truthful position, seems like a very good tell.

Not sure you can do this in one question. But if you know the in-group of the person, probably the best thing to ask about is someone your from that in-group hates or is supposed to hate.

So if they are liberal, ask them about Trump. If they are conservative, ask about Hillary. If they are a baseball fan, ask about the Yankees. If they like the NBA, ask about the Warriors (who can obviously do no wrong, so please don't even bother attempting to deny the perfection of the Warriors or I'll plug my ears, and I'll continue to believe this is a perfectly rational position nah nah nah nah).

Does a rational person align in the first place with groups as varied (and internally contradictory) as "liberals" and "conservatives?"

Haha, I hope they do get him fired.

I wonder where the word "rational" even stands. We know that emotion pervades life, and that healthy emotional balance is necessary both to define goals, and to seek them effectively. On the other hand we see emotion blocking progress all around us, either by confusing goals or making ineffective plans for attainment.

By those constraints, finding something you like on the menu, and communicating it effectively is rational. Accepting a bad choice with good grace is probably rational too.

I am not sure about upgrading on "considered and uncertain and complex" answers. Those might be smart, but they might show an excess attachment to novelty over truth.

I think this is a lesson in futility. The problem is, there are so many confounding factors involved--both in the reasons behind the person's response, and in your interpretation of their response--that the whole thing would be nothing more than an exercise in confirmation bias.

Take the first two questions that I saw in this thread. "Would you be happy to receive an income tax return?" I acknowledge that the money could be put to better use than an interest-free loan to the government, but the benefits of having a large amount of my income being untouchable until a set time assist in budgeting and it reduces the hassle of managing the taxes in other ways (no small thing given my lifestyle). This is a complicated issue, getting into work, family life, political philosophy, etc--not something easily assessed!

Similarly, "Did you vote for Hillary?" I know many people from various political philosophies, ranging from socialism to Objectivism, who voted for her. I also know people from those same backgrounds who refused. Their reasons are complex, again getting into a myriad of philosophical issues--not something you can use as a single question.

Rationality isn't about answers, it's a process. Outside of extreme cases you cannot understand a person's process via a single interaction. If they're wearing a tin-foil hat and ranting about aliens replacing humans, yeah, you can say "They're nuts". But ordinary people? The issue is too subtle to allow for the application of simplistic heuristics.

It comes down to something very simple. "I'm Rational and you are not!".

"having a large amount of my income being untouchable until a set time assist in budgeting and it reduces the hassle of managing the taxes in other ways (no small thing given my lifestyle). This is a complicated issue, getting into work, family life, political philosophy, etc–not something easily assessed! "

To this point, check out the Tanda system in Mexico and some Mexican-American communities. It's essentially a lending circle in which a different person receives money from everyone else every month. In a country with a reliable, low-barrier, interest-giving banking system like the US this system is irrational, but in an unreliable, high-barrier system like that of rural Mexico in which hoarding cash in a house is also risky, a tanda system solves a lot of systemic problems at once.

I'd ask "what do you think of Trump?" or some other question about a controversial topic that tends to elicit emotional responses.

The important thing isn't so much the answer, but how they answer. In a less rational person, the response will be laden with simple and reflexive notions, and probably quite a bit of emotion. In a more rational person, the answer will be thoughtful (even if vigorous) and will be likely to treat the question from a number of angles.

By your standard, Trump himself is irrational.

No. That is the way to find a liar. Someone who blathers on about rational reasons to do this or that in politics is simply flattering themselves. The person who blurts out some reflexive notion is being honest and frankly saving you and the universe some time to focus on other more interesting and important things.

A more interesting question is who they thought was going to win. That leads to a very interesting train of thought about who we listen to, how aware we are of the concerns of our fellow citizens, whether we are open to challenge of our priors.

That triggers a quick objectivity question: how could Trump win when odds were 70:30 against?

A 30% likelyhood event isn't that unlikely - it happens once out of every three times. But, the presidential election is just one trial. Run the same election with the same conditions 100 times (impossible...) and Trump would have won 30 times.

But you should be a Bayesian and update your priors since Trump actually did win the election...

I agree Chris, but an amazing number of people settled it to a certainty, and then a proof/disproof of that certainty.

"That leads to a very interesting train of thought about who we listen to, how aware we are of the concerns of our fellow citizens"
We have listened about the silent majority and the "concern of our fellow citizens" for decades - even I, a non-American have. Why now? Why Trump?

Definitely. Let's just have a king instead. This is all soap opera.

Well, clearly the answer is to ask the questions that lead the blog post. :)

My answer to "who he or she thinks are the rational people" would be "nobody." But does that make me more or less rational than the mean person? And are the mean and median person different? Why?

Is the scientific method that necessary, or even useful, today? Sure, in times past, when superstition seemed reasonable, a fortune cookie contained one's fortune, and “omens” portended awaiting catastrophe, but today? What's your defense of the scientific method today? [Disclosure: I "borrowed" this question.]

Are there still people who oppose vaccination? Then, yes, it's necessary.

This presupposes that everyone has had the scientific method explained to them. Sadly many have not, and potentially rational people are left without a valuable tool.

Full disclosure: Cowen's post brought to mind My Dinner With Andre, and the subject of the scientific method was one of the last discussed by Andre and Wally.

The purpose of science and the scientific method is not to discover universal truths and enlighten our species but rather to create models that allow us to correctly predict interaction of the natural world or between human actions and their effects. So the scientific method is a form of an iterated model improvement function, and to that end is as useful as it has ever been.

I would ask someone to tell me a joke that they heard recently. Rational people laugh appropriately. Irrational people don't.

Germans - or basically anyone not from your culture - are the ones acting irrationally by not laughing when they are told something that the other person believes a joke?

Rational Germans would realize that laughing at a joke is a social signal as much as genuine humor, and would react accordingly in a meal with a relative stranger.

If you said, "Bitte, sagen Sie mir ein Witz" I'd expect everyone to at least chuckle at the resulting joke. If you said "Sagst du mir ein Witz" I would expect a few people to express confusion over the joke and not laugh.

'Rational Germans would realize that laughing at a joke'

You keep thinking that what is clearly a joke in one language/cultural context is universally recognized as a joke by everyone from any of the other existing languages/cultural contexts. This is simply incorrect.

'If you said, “Bitte, sagen Sie mir ein Witz” I’d expect everyone to at least chuckle at the resulting joke.'

If the joke involved a pun, required a shared cultural reference (a Swiss joke concerning a Swiss soccer coach is unlikely to have much impact to Austrians or Germans), or even just assumes a certain amount of shared knowledge to recognize a punchline (hard as this might be to imagine, basically no other English speakers but North Americans know much about baseball, to put it mildly), it is quite likely to not even be recognized as a joke.

Politeness is its own discussion, of course, but a typical German tends to be blunt, and is as somewhat as likely to say that what was just said was stupid as they would be to laugh. Culture plays a huge role in humor, though I'm sure many of the most loyal commenters here would argue that humor has nothing to do with nurture, and is clearly genetic.

What's shortest book in history?

German Humor!

If smoking is so bad for you how come it cures salmon?

Well, that was a much more succinct example of a joke that would need to be explained to anyone who does not speak English well. As noted by google translate using German as a concrete example - 'Wenn das Rauchen so schlecht für Sie ist, wie kommt es, dass es Lachs heilt?' And of course, 'curing' is more or less translated as 'räuchern' as smoked, or 'pökeln,' neither of which create a pun in German. Plus, broadly speaking, puns are not something Germans seem to even be that aware of, much less aware of a way to joke. Whether someone would find that translated sentence intrinsically funny on its own is another discussion.

Isn't there a semi famous German pun in Das Leben der Andersen about a telephone?

Anderen. Thanks autocorrect.

Sounds like you like fake people.

Rationality works along lots of different axes. E.g. I could imagine someone otherwise highly rational who say, had a gambling addiction or believed in a religion that was apparently rather irrational in nature.

As we're eating, I'd choose to look at rationality qua food.

1. An epicurean approach. Do they pay attention to what they are consuming? Is it important to them? Good food is one of the greatest and most constant pleasures that life has to offer, and to be indifferent to that is hardly rational.

2. "Do you like pizza? Why?" If you've never managed to formulate an opinion about pizza and have trouble justifying it I have doubts as to your rationality.

I will tell you a question I purposefully do not ask. If someone tells me their goal, I will not ask "how will you achieve that?"

It too easily lays bare the irrationality of the human condition.

Or lack of information about the future.

Would you pay a premium for the hype/social value of a product in lieu of getting a cheaper one functionally on pair?

I carried a lot of books around with me during law school, which turned out to be probably among the least rational things I've done in my life. I suppose the books had more use than 50 pounds of dumbells that I could have carried around instead.

I say "turned" out, but was probably acting irrationally at the time - not with the benefit of hindsight.


You would do the marshmallow test.

You can have one marshmallow now,

And, if you can wait.

You can have two 15 minutes from now.

But I want the marshmallows now! What good they will do me 15 minutws from now? Future me will not be able to enjoy them as much as I would.

Or you could ask a female dining partner if she has a marshmallow in her panties - then you would know she is pledging (or whatever term is correct - Mason was thankfully pretty much greek free) a sorority at a mid-western public university. You would then also know what she will be eating for dessert later - well, at least among her fellow pledging sorority sisters. Further proof of irrationality is not required. Of course, asking that question could be a bit problematic in terms of rationality, too.

This presupposed that rationality begets willpower, when there is probably no correlation between the two.

Irrational people are generally impulsive people who have a binary picture of the world.

So pick a hot button, emotion-laden topic and ask if they can see both positives and negatives.

Ie, Trump vs Clinton, abortion, global warming, immigration.

Shorter version: ask them when is the last time they changed their mind on something significant.

Upvote for the question of last time you changed your mind on something significant, or how often you change your mind.

I'm going to start using this as an interview question.

If I were your interviewee, I would interpret that as, "When was the last time you came around to what is most likely to be my point of view?" But you could give points for correctly discerning that. It would be a foolish person who would betray their thoughts on anything significant and unrelated to the expected task, in a job interview - especially given that in many workplaces, they can easily get in trouble for expressing a point of view, or even being overheard discussing, however mildly, any "divisive" topic.

Rational Actor (RA) is attempting to discern the rationality of Dinner Companion (DC) at a restaurant table first by not asking any direct question requiring verbal response.

RA only has to see what DC orders: is DC ordering food far in excess of his RDA caloric intake? foods excessively high in fat content or bad cholesterol? alcohol to excess? (how many glasses get knocked over during the course of the meal, how much silverware gets dropped during the meal?) orders of meat dishes exclusively? These can be deemed reliable indicators themselves, since at this stage verbal responses have been deemed out-of-bounds.

When the time comes for RA to pose a question directly to DC to discern rational capacity, RA needs merely to pose the question: "you're picking up the tab here, right?" (This question might have greater value if RA has dined exclusively on expensive menu items.)

Observe their engagement with the menu. If they're quick to find reliably tasty food to refuel their cellular engines they're rational. If they scrutinize the menu in search of the optimal unicorn food-like experience they're irrational.

I think that is shaped by blood glucose levels as much as character.

If it is a nice restaurant, or one that the guest is unlikely to visit again, spending more time on the menu is more rational as the opportunity cost of ordering a less than optimal meal is higher.

"How well they treat the server would be another variable of interest."

Are you confusing 'rational' with 'nice' or 'decent', here? I think this is a very good test, but not for rationality.

Rationally, the server is a resource to be used effectively. Being nice and decent isn't charity. It's win-win.

If the server fears being beaten again, he will be much more solicitous and helpful.

"Double or nothing on the tip?"

This allows you to assess risk aversion on small stakes (irrational) and, with a 50% probability, assessment of tipping behavior, though I admit that the latter, while an interesting exploration of rationality, is not directly interpretable without conversation and inference from verbal responses.

Credit card roulette!

This is definitely true.

Non-verbal cues: assuming you are rational (and you must be to assess this in the first place), are they drifting off when you talk about your super-smart rational thoughts, or are they focused, engaged, smiling, furrowing their brow, etc? Rational people seem to like hearing smart thoughts, deep thoughts, puzzles, etc. Less rational people like hearing sportsballs and Kim's manicure. Their reaction to both sends a pretty clear signal.

Question: Something to test availability heuristic that's non-controversial. "Do you think traffic has really increased here recently" or "I read a lot of reports about delays at the airport, do you think that's gone up recently?" Politically charged things, well, people can rationally check out of that, or rationally irrationally root for their own time (which is the actual benefit of politics anyways). Non-political things that everyone has experienced....well, that's a better option.

Views on race and IQ. In no other field is the evidence and common sense so overwhelming, but conventional wisdom so wrong. If a person brushes off the possibility that races differ in IQ you know he's incapable of an independent thought that would be out of place on the NYT editorial page.

That is a sad failure. You may leave the lunch.

A rational person would be talking about regional populations, and would understand at the outset that they do not map neatly into the American cultural definition of race.

People put Mayans and Iroquois in one bucket because they are not firing on all rational cylinders.

Exactly! That's why there's so much overlap between scores on IQ tests given in East Asia and those taken in Sub-Sahara Africa. Cause our "cultural conceptions of race" based on what people look like obviously has no connection with reality.

This kind of common sense approach to the question is the essence of rationality. Nonsense like this is the opposite, and that's why race and IQ works so well. This isn't very difficult.

This is a pathetically flawed argument. Spare half a moment to explore other correlated and conflating variables: cultural attitudes toward education; average incomes; caloric intake; institutional stability. Attempting to extract a parsimonious explanation out of such an obviously multicausal situation is little more than drivel and you are embarrassing yourself.

Desperate sophistry. If you've read anything on the matter, you would know that "caloric intake" and "average incomes" aren't the answer, as Taiwan and China were scoring equal to Westerners back when their incomes and living standards were a fraction of modern day Africans and blacks in western countries.

You only need assume a "multicasual situation" when one variable can't explain what we see.

*eye roll* use of adverbs and the term "parsimonious" are excellent cues to your irrationality.

How convenient for you that "*eye roll*" escapes your disdain.

I see what you did there. You did not stand on "race," instead you shifted a bit toward historic populations. If you were not aware of your mind trick, you might be irrational and not merely dishonest.

Specific historic populations in Asia and Africa are quite varied.

Along those lines, I will give you some real population genetics to irrationally ignore.

Yup, typically denier. Play word games when you have nothing else.

You are unable to process my answer or how it relates to the link.

I should not be surprised. If you could read and comprehend modern population genetics, you would not be stuck where you are.

.. reinforcing the idea that human populations are far from homogeneous, are highly intermixed, and do not fall into distinct races or castes that can be defined genetically.

You're even worse than the typical denier, you're still on race doesn't exist. The smarter ones have moved on to arguing about the Flynn effect, the last thread you have.

1) Do you think that a person who believes that individuals with high skin melanin count are inferior, will be more or less likely to hire them?

2) Speaking to your open mindedness, is it possible that the answer to 1) is related to observed differences in income?

3) How hard do you usually try on tests given to you that a) have nothing to do with your life and b) you don't really want to do anyways?

I think checking whether they are carrying a book is more likely to detect Jehova's Witnesses or other missionaries.

Think observing in the midst of a semi-stressful situation is a good indicate. Ie your plane leaves out of JFK in two hours, how do you manage the process.

1. I'm going to go with "Do they order food that they can somewhat easily make at home?" Ideally I would put them in charge of ordering shared side dishes for the table and then see what percentage of the side dishes (less any that we mutually agreed weren't good) were eaten conditional on my appetite being satiated, but there are a lot of complications to that test.

2. Asking about Trump is really not that bad, but some people will have trained themselves to give a brief, polite, nonconfrontational answer rather than an honest one- including rational people. "Is there anything big you're saving up money for?" seems like one angle to take as it would tend to elicit irrational answers from lying irrational people.

“Do they order food that they can somewhat easily make at home?”

I can easily make steak or pasta at home. Why is it irrational to do so at a restaurant if I enjoy eating steak or pasta?

I would wager that steak or pasta are in the top 3 of what people order when they dine out.

Why conflate speed of decision-making and rationality? I think speed is self-referencing bias for some people, not an indicator of quality of thought.


Make bad decisions faster.

Of course I'm not rational. If you divide me, there will always be a remainder.

But do you enjoy pi?

To me, it's obvious that you and your companion should trade at least a bite of what's on your plates, unless there's an active aversion to what the other person has ordered.
At lunch with friends, each ordered a different "tartine": egg salad-with-roasted-tomato, goat-cheese-and-salmon, brie-and-jam, tuna-and-capers. The sandwiches there come without chips, just a thin slice of out-of-season melon, and a couple of berries. So you're eating a whole lot of brie, or egg salad. I proposed we trade, and all seemed unusually happy with the result, but I don't think they would have initiated it.

I'm inclined to think you can't tell much about someone's rationality by dining habits, because they will be heavily biased by class.

For one thing food choices are often a complex calculation that depends on price and the person's disposable income. A poor person is going to have a harder time selecting a menu item because if he chooses wrongly and doesn't like it, then he's wasted a larger chunk of his disposable income than a richer person.
I remember this distinctly as a difference from my student days to my professional upper-quintile income today. I would worry a lot more about ordering something new and not liking it.

Napkin use and such - do you eat the bread, do you use your fingers, do you order water or a soda - are similarly class indicators, not intelligence indicators.

If pressed, the one thing I would say is, do they use the bathroom before eating. A rational person will order first, then use the washroom and wash their hands before the food gets there, because it's a more efficient use of time.

I agree, and eating at mostly familiar places, ordering something you know you like and that will not leave you needing to fix something else when you get home, is for me, who must cook every day, a perfectly rational strategy.

"What do you believe that you think might be false?"

That's a great question.

That I may be fallible.

Says the person with the most spittle flecked unupdated priors on the internet. It takes 30 seconds of watching you whipping yourself into an orgy of hatred against the WWC to know rationality just ain't your forte.

You just wanted to say "orgy" while saying mean things to women.

Which is kind of messed up.

begs the question "whose rationality". Before one can really answer the question one needs to understand the actors understanding of the world around them (resources) and their own set of goals and priorities across those goals.

I find it helps to ask people their opinion of what other people think. Pick a 'rational' subject to discuss, ask them what others think. People can be resistant to discussing controversial or irrational opinions, however anchoring & confirmation bias usually means that the same individuals find it very difficult to conceptualize that others could hold opinions that differ in any significant way from the ones that they themselves hold. Or maybe I'm being irrational in holding to this belief?

The least rational thing I've done this week is read these comments,

Oh hell yes, Ron!
I drive a forklift to pay my bills and wanted to see what people who read a smart blog would think think of as rational, 'cos maybe they'd be at least self educated but what's this?
200 Dimwit Will Hunting LARPers, 4 good calls, 3 laugh-out-loud jokes and a new troll response for the What Is Love? gag that some people online seem to think is still funny.
Virtue signalling and teacher's pet suck ups hitting an energy level over 9000.
But at least I learned that the Philistine army's super-soldier who was being relied on to do an Achilles deed and single-handedly win conquest of an entire nation was slow moving, cross eyed and myopic because of his tumor, so it's not a total loss.
Bless your heart Malcolm "The Forehead" Gladwell. You're a treasure.

For #1, just a few:
- seasoning with pepper/salt before tasting the food (I see it often)
- eating very slowly, to the point where food is cold, but continuing to finish plate
- tasting wine without letting it breathe first.

I remember reading that Adm. Rickover used the salting food before tasting test.

"tasting wine without letting it breathe first"

The horror!

You are mistaking snobbery for rationality.

Let them choose a seat first. Do they choose a seat with the back toward a wall? Are they facing the door? Is it possible they're wearing a shoulder holster?

"Would you please name some high profile people you consider rational and a few that you consider irrational and why?"

At minimum, you would get some insight into if the person understood the concept of rationality.

So much for my reading skills. This was one of Tyler's suggestions.

Do you think Tyler is losing it?

Tyler is playing 5D chess.

I wonder if 118 comments (so far) indicates rational responses to an idiotic blog post.

this entire test premise is a flirting dangerously close to a fundamental attribution error. why should we believe that rationality is even a trait of people? it seems more like a learned cognitive modality that has some utility in certain contexts but not in others.

the real issue, however, is believing that there are consistent signals of rationality at all. at best, this test asks for superificial social signals that people who are in the same social class as Tyler might use to signal to each other that they have learned similar or compatible cognitive modalities.

Rationality is subjective and relative to perspective and goals, conscious and unconscious.

We're not machines so it is simple enough, no one is truly rational.

Do you believe that their is some magical elixir that exempts labor from the supply and demand nexus?

Do you believe you can make a country more libertarians by flooding it with people from countries completely hostile to libertarianism?

By these standards Tarroback and Cowen are some of the craziest, most irrational people on earth.

Do you think you can make a country more libertarian by electing a national socialist?

Ask the dining companion to sing a melody and listen for the consonance of the intervals.

Do they prefer the magic of David Blaine or Chris Angel?

It's saying a lot that this isn't the silliest suggestion on here.

See if they fall for the if your hand is bigger than your face your have cancer trick.

Alternatively you could pants them and see how composed they act.

Go to a buffet. Watch what they eat.

This. This is what great internet entertainment is made of.

If you define rationality as the use of relatively better methods for problem solving, then in both cases you want to face your counterpart with a problem.

In the first case, you should use your environment. For example, if you are at a restaurant that the other person is visiting for the first time, the menu choice is a perfect opportunity because he/she doesn't want to waste your time while deciding, but on the other hand he/she a priori lacks the information to choose. A good trade-off. Therefore, you could look at how he/she gathers information: does he/she ask you, risking your limited knowledge of the menu, or does he/she ask the waiter, risking the effects of incentives towards recommending more expensive plates? Does he/she research on his/her own, looking at clues of quality in the menu design (plates at the top/front, pricing, etc.), the restaurant's own signaling (is it italian food oriented?), etc. Or does he/she simply copy your choice?

For the second part of the problem, you need to ask an impossible question, where you don't have a true answer. At the risk of cheating (or maybe this is intentional?), I think a good exercise would be using the question you are asking. Not the one about defining rational people, but the one about what would you ask to assess a person's rationality, where you can ponder the aproach of the respondant by his/her use of definitions, assumptions, consideration of biases, and, in general, how this person identifies relevant factors at the moment of defining the problem.

You raise some good points. If the purpose is to probe the psyche of the subject, rather than to eat a meal, a great advantage can be had by enlisting the collaboration of the waitstaff. How does the subject react when the waiter "accidently" dumps a bowl of gazpacho in his lap? How does the subject react when the waiter takes great offense to a very slightly off-color remark by Tyler? A $20 bonus in advance should ensure great performances by the waitstaff.

Then, one waiter shot the other with the fake gun. Unperturbed, Tyler remarked "I've never seen that before." Subject seemed anxious but not panicked.

Offer to pay for their meal and then proceed to insult them throughout the meal. If they stay until the end they are rational. If they get up and leave they are irrational, but also human. Also this means you are an neoclassical economist and this is a great example of why you have no friends.

During the peak of interest in the use of lobotomy as a therapy for mental illness, one of the problems was there wasn't a good test to distinguish lobotomized from non-lobotomized people. IQ tests did not work because often measured IQ would increase after a lobotomy, probably due to an increased ability to focus on taking the test. The test with the highest sensitivity was the Robinson test, which consists of writing the words United States of America in cursive writing as slowly as possible without stopping. The highest scores were from teenagers and young adults, who could take many minutes to complete the test. Lobotomized subjects had the worst scores -- they would start writing slowly, but then break out into their normal writing speed.

Were a respondent to Tyler's question to recast the matter as one of wisdom and virtue, that would be a good sign that she is wise and virtuous.

I like Peter Thiel's "Tell me something that is true, but nobody else agrees with". I'm not sure if this tests rationality so much as it tests a person's contrarian nature or their ability to think outside the box. But maybe.

> If you were trying to assess a person’s rationality on the basis of one not-directly-verbal indicator, given his or her behavior over the course of a meal, what would it be?

Put my phone in airplane mode. pretend to be upset and complain that it doesn't work, hand it to them and ask for help.

If they pretty quickly figure out how to switch it back into normal mode, they're a rational person. If they spend too much time guessing at improbable dead ends, then, they're not so rational.
Navigating menus and other man-made systems is simple process that requires only attention, patience, and a decision making process that isn't pushed and pulled around by irrelevant data.

Are you waiting for a call from a client or something?

If not, I'm not interested in your phone problems, but notice that your world is falling apart without your crutch.

1. Let them sit down first, but you continue to stand. Assess their reaction.

And then make a grimace like you would expect from PM Abe if one of his aides sat down before the Emporer did. Assess reaction.

I know someone who carries books in order (I suspect) to seem sophisticated. Their life as a whole is a failure. Their attending to books rather than putting their life in order is quite irrational.

"The pre-eminence of man over beast is naught, for all is הבל" (Ecclesiastes 3:19)

Non-verbal test: this is old and not original to me> observe if they season their food before tasting it.

One of my favorite applicant questions when I was interviewing job candidates was to ask them if they ran the DC Metro and a man had a heart attack would they order that the train remain stopped until the paramedics could cart him off. Everyone would answer yes to the first step in my multi-part question. Then I would add other factors to see how they reasoned the cost benefit analysis. Eventually, the entire system would have to be stopped for months because the person could not be moved, and he had only a 1% chance of survival, etc. Most people bailed in the middle of the add-ons, but one young man stuck with shutting down the system for a lengthy period even when the chances of success were infinitesimally small.

And if you could ask only one question of a person, to assess his or her rationality, which question would it be?

Do you think it's possible to assess someone's rationality by asking just one question?

Why are you worried about rationality? Isn't a a rational person what leads to complacency? For example, wouldn't a person who grew up in a small town in Montana, with proven friends, a proven home-town industry to provide a solid living (oil or railroad), and proven thoroughly enjoyed hobbies (hunting, fishing, skiing) want to stay there and maximize their utility doing things they know will provide that utility?

It seems to me slightly irrational to leave behind what will guarantee one utility for a chance of getting that utility back somewhere else, doing non-complacent things.

Fascinating question, though I agree with many commenters that the exercise may ultimately be futile.

But of course I will have a go at the second question (verbal),on the building on things Carl and M Stacey said.

I would ask "What is the purpose of this dinner, and, given your answer, what should we do?"

If the person says "To enjoy the food," then the follow on is something like "So, let's eat, and only indulge in small talk, since any weighty topic would require long spans of time spent talking, and the food is best eaten fresh and hot."

If the person says "To have a fascinating conversation," then the follow on may be something like "So, let's save a lot of money and time, and instead of renting this high-end conversation booth and constantly interrupting our trains of thought with chewing and choosing, let's adjourn to a cheap coffee shop, have our talk in 30 minutes, and reclaim the rest of our night and a couple of hundred bucks."

(One can hypothesize other answers, such as "To suck up to TC to get a good letter of recommendation," and then hypothesize rational courses of action that ensue.)

I laughed heartily at this thread. In case anyone is still reading, the obvious answer to a) is to observe whether your server identifies your companion as "sir" or "ma'am."

How about the indicators:
1. Observe if the person appears to believe he can assess the rationality of a stranger over the course of a single meal via observation.
2. "What do you know that you know?" (An alternate form is "What do you know that you don't know?", but for a rational person, that should take much longer than a single meal to answer)

I would ask if the person spits or swallow. That tells you right away. Tyler seems like he swallows.

2: Describe a time you've been convinced to change your mind about something important to you.

Also, I think that is the best option and no one could convince me otherwise.

Second one is very easy. Ask them for what their idea of God is?

My main takeaway from this is that I'm going to be an order of magnitude more concerned about the potential for intense scrutiny by my dining companions. Apparently the wine, the salt, the napkin, the menu, my relations with the server, the food I order, the rate at which I consume it might indict more than just my table manners and social skills. Which raises the question: are things like table manners and social skills "rational" in form, or only in terms of the expected outcome of their correct application?

I wonder if all the commentators here were seated at the same table if any would come away with the impression that any of the others was rational.

I personally suggest the old chestnut "Would you rather fight one horse-sized duck, or 100 duck-sized horses?" Anyone who thinks they can take on a giant duck is not thinking rationally. Obama, disappointingly, chose the giant duck.

My main takeaway from this is that I'm going to be an order of magnitude more concerned about the potential for intense scrutiny by my dining companions. Apparently the wine, the salt, the napkin, the menu, my relations with the server, the food I order, the rate at which I consume it, and whether or not I want dessert might indict more than just my table manners and social skills. Which raises the question: are things like table manners and social skills "rational" in form, or only in terms of the expected outcome of their correct application?

I wonder if all the commentators here were seated at the same table if any would come away with the impression that any of the others was rational.

I personally suggest the old chestnut "Would you rather fight one horse-sized duck, or 100 duck-sized horses?" Anyone who thinks they can take on a giant duck is not thinking rationally. Obama, disappointingly, chose the giant duck.

"How well they treat the server would be another variable of interest."

If the assumption is that a rational person treats a server well, I do not follow. Is this a restaurant you plan to frequent over and over? Then yes, some long-term relationship with the staff may lead to better outcomes over time.

Are you traveling to a place you will likely never eat in again? A rational person pays little attention to the server, demands the maximum amount of service short of getting food spit in or dropped on the floor, tips $0, then leaves. Maybe that's a little extreme, there is some joy in being a kind person (or am I just irrational?) but putting energy into being nice to a server that you won't develop a relationship with doesn't make sense.

"The intuitive mind is a sacred gift and the rational mind is a faithful servant. We have created a society that honors the servant and has forgotten the gift.

– Albert Einstein"

Are their thought processes servant like?

Ask what is a great human gift.

As others have noted, rationality is not a fixed trait. The likelihood of a person thinking rationally during a particular meal is heavily influenced by his physical and mental state. So I would be looking for non-verbal indicators of good health, relaxation, alertness, and a neutral to good mood. Negative indicators would be sniffles, yawns, furrowed brows, fidgeting, clumsiness, the smell of alcohol, lack of eye contact, or excessive physical contact. If I were allowed to ask a question I'd just want to know "how's your day going?"

Comments for this post are closed