Smart people are especially prone to stupid mistakes

Jonah Lehrer takes to The New Yorker to discuss Thinking, Fast and Slow, the latest book from Daniel Kahneman, a psychologist who's won the Nobel prize in economics. Lehrer discusses Kahneman's contention that smart people are no less prone to cognitive bias than anyone else, but are prone to believing that they are immune to error. Kahneman himself admits that he makes systematic cognitive errors all the time, even though he's devoted his career to studying them.

This has particularly grim implications for a society that thinks it is a meritocracy but is really an oligarchy, because the competitively educated people at the top believe (incorrectly) that they don't need to have their intuitions reviewed by lesser mortals.

And here’s the upsetting punch line: intelligence seems to make things worse. The scientists gave the students four measures of “cognitive sophistication.” As they report in the paper, all four of the measures showed positive correlations, “indicating that more cognitively sophisticated participants showed larger bias blind spots.” This trend held for many of the specific biases, indicating that smarter people (at least as measured by S.A.T. scores) and those more likely to engage in deliberation were slightly more vulnerable to common mental mistakes. Education also isn’t a savior; as Kahneman and Shane Frederick first noted many years ago, more than fifty per cent of students at Harvard, Princeton, and M.I.T. gave the incorrect answer to the bat-and-ball question.

What explains this result? One provocative hypothesis is that the bias blind spot arises because of a mismatch between how we evaluate others and how we evaluate ourselves. When considering the irrational choices of a stranger, for instance, we are forced to rely on behavioral information; we see their biases from the outside, which allows us to glimpse their systematic thinking errors. However, when assessing our own bad choices, we tend to engage in elaborate introspection. We scrutinize our motivations and search for relevant reasons; we lament our mistakes to therapists and ruminate on the beliefs that led us astray.

The problem with this introspective approach is that the driving forces behind biases—the root causes of our irrationality—are largely unconscious, which means they remain invisible to self-analysis and impermeable to intelligence. In fact, introspection can actually compound the error, blinding us to those primal processes responsible for many of our everyday failings. We spin eloquent stories, but these stories miss the point. The more we attempt to know ourselves, the less we actually understand.

Why Smart People Are Stupid


  1. So does he think he’s stupid or is his theory wrong?  (it’s a fair bet that Nassim Nicholas Taleb or Malcolm Gladwell would say both)

    1. Both this article and the one yesterday about meritocracies confuse intelligence with arrogance (“…the competitively educated people at the top believe (incorrectly) that they don’t need to have their intuitions reviewed by lesser mortals”)

      The key is not to fail to reward merit, but to have the courage to stand up to the arrogance of others, more intelligent or otherwise.

      1. smart people drink more:

        1. If you can see most of the probable outcomes in any given situation, you need a wee nip to take the edge off.

  2. “smarter people (at least as measured by S.A.T. scores)”

    There’s your problem right there. SAT’s don’t test how smart you are. And even if they might be a reasonable measure of intelligence, the kind of test prep and score-inflation many people undergo to get high scores on the test is more akin to gaming the testing system than actually being smarter at a certain intellectual skill (focusing on process of elimination instead of picking the best answer, for instance). So a lot of people with astronomical scores on the SAT are actually quite stupid and instead got good tutoring or prep, while a lot of really smart kids take the SAT and tank it (especially if they’re in underserviced neighborhoods and don’t have the same access to test prep).

    So if we change the thesis from “smart people” to “people who think they are smart because they scored high on the SAT and went to good schools as a result” then yeah, I can totally see the blind spots in action. I see them CONSTANTLY.

    1.  And one obvious solution is to have a rotating selection of other people check your work and point out flaws. My co-workers and I often listen to each other’s conversations and pipe up with a correction about out of date information or a new process, to which the standard answer is “ah yes, I’d forgotten that”, followed by confirming that information to the other person.

    2. This is so true. What I take away from this is standardized tests aren’t good measures of actual intelligence, and getting into a great school doesn’t mean you’re that smart or particularly useful–but you may still be neurotic.

  3. There are a number of things getting tied together (perhaps unfairly).

    1) Yeah, smart people have cognitive biases, (probably reinforced by the same type of subconscious response that makes them “smart” in the first place – knowing what to dismiss).

    2) Almost everyone will regard good fortune as “deserved” rather than “luck”.

    3) Meritocracy moves to Oligarchy by making sure your kids get the “chance” they “deserve”.

  4. Best way I’ve found to avoid making dumb mistakes is to always remind myself of this quote from Mark Twain,  “As for me, all I know is that I know nothing.” 

  5. Every six weeks or so I’ll be very proud of myself for knowing a slick way to get to Loreen Dr or Lumley Rd when I actually need to go to Luann Ln or Langley Ln.

  6. I think we can simplify this a little. Change “smart people have cognitive biases” to “all people have cognitive biases” and you’ll get to the heart of the matter.

    1. I think you are absolutely correct here DW. But there are at least a couple of things that can make things worse for intelligent and educated people:
      1. They are more adept at defending and pursuing ideas – so they can dig themselves in deeper.
      2. They don’t always subject their preferred verification criteria (peer review; scientific methodology; what they say in the NYT etc.) to sufficient scrutiny. The danger here is that if errors can get past these filters they become received wisdom and defended as orthodoxy.

  7. I predict a debunking for this one.  Partial debunking at least.

    An actual education should increase self-questioning the constant re-evaluation of believed truths.  Any educator who leaves you with the impression that you’re too clever for doubt was a pretty piss poor educator.

    Hell, even just good test taking skills should have fixed the bat&ball problem, because the last step should always be to validate your answer, because validation is generally much faster than coming to an answer in the first place, and most tests give you plenty of time to check your work.

    I dunno.  I guess time and more research will validate or debunk this one.  Or pare it down into something less headline-worthy.

      1.  And with the Baseball problem, why check my answer if I the only person who’s going to know I got it wrong is me? I’m much more likely to actually use math if the outcome has some kind of actual risk/reward component.

  8. Go go Checklist Manifesto! Have the humility to know when it’s critical to know you’re fallible!
    On the other hand, doesn’t this contradict just a bit the results found in what has to be my favorite psychological study of all time, entitled Unskilled and Unaware of It?

    From the abstract…
    “People tend to hold overly favorable views of their abilities in many social and intellectual domains. The authors suggest that this overestimation occurs, in part, because people who are unskilled in these domains suffer a dual burden: Not only do these people reach erroneous conclusions and make unfortunate choices, but their incompetence robs them of the metacognitive ability to realize it. Across 4 studies, the authors found that participants scoring in the bottom quartile on tests of humor, grammar, and logic grossly overestimated their test performance and ability. Although their test scores put them in the 12th percentile, they estimated themselves to be in the 62nd. Several analyses linked this miscalibration to deficits in metacognitive skill, or the capacity to distinguish accuracy from error. Paradoxically, improving the skills of the participants, and thus increasing their metacognitive competence, helped them recognize the limitations of their abilities.”

      1. And why the majority of the people behind the wheel figure they are better than average drivers.

    1. I’m not sure it does – if you have a prejudicial goal outside of the research. These pieces combine into a perfect one-two punch if you’ve got an engineer in your life who has opinions on things that you need to shut up now.

  9. I was at AIG when the dust settled on that disastrous spreadsheet app that a few geniuses in England had created and sold to exec row with no input from anyone in btwn. Talked with some geeks in IT about it one day and everyone said the same thing: if this thing had been subject to technical review outside the inner circles, it would never have gotten to 1st base, it was that bad. That app was the spade that dug the $300B hole that US taxpayers are to this day still filling. And it was allowed to make that hole because it was an insular game played btwn. “the smartest guys in the room”, with not a single even semi-independent review. 

    It’s the same phenomenon in the individual self: intellect pushed onto the stage of life naked and alone becomes a demon, a fool, a rabid dog. More on that part of it here:

  10. When I was a college freshman studying design I realized there were 3 kinds of people: 

    The rare ones who actually know exactly what they’re doing.

    The nearly as rare who realize they don’t know what they’re doing.

    The masses who have no idea that they don’t know what they’re doing, and plow on anyway and get rewarded, because their instructors don’t know what they’re doing either.

    1. Is there a 4th category (where I think I fit) which includes all 3 of yours at various times and in various subjects?

  11. Actually, you have it *ALL* wrong.
    You see…the issue here is that stupid people are accustomed to doing stupid things, so when they do do something stupid, they recover quickly since they are so used to doing stupid things. The problem is when an intelligent person does something stupid. They are not used to this so it usually is catastrophic. 
    So it will appear that when intelligent people do stupid things, that they do more stupid things than stupid people, but that is not the case. When a smart person does something stupid, it becomes more of an issue because the intelligent person has difficulty recovering from the stupidity, as such, it can snowball into a real big problem…

    1. Hmm. Not sure that smart stupidity is inherently more unstupidable than stupid stupidity. The smartness behind smart stupidity’s unstupidability may be sufficient to overcome the sluggishness of the unstupidality of that smart stupidity’s stupendousness of which you speak. There’s probably, ultimately, a balance of stupid.

  12. At Landmark forum trainings (EST reincarnation), the one thing that seems to blow everyone’s mind is the “things you don’t know you don’t know” quadrant.  I’m ashamed to admit it was an unfamiliar concept to my ivy-league-educated self. 

    But I’m lucky; my brain has a quirk of making stupid mistakes like typos or misspelling words I know how to spell, much more often than other people.  This forces me to constantly be “checking my work”, and so I tend to recover from my mistaes.

  13. This assumes the people at the top are smart and talented, which is almost never the case.

  14. Perhaps a system of screening for those biases and mistakes could be invented. In it, we could demand that assertions be examined and tested independently, that facts be confirmed by repeated experiments… but what might we call such a system? Buddy Check?

  15. I assume, while being a ‘smart person’, I can and will make mistakes. I count that as one of the things that MAKES me smart; acknowledgement of mistakes and endeavoring in not repeating them…

        1. Be careful!   I’ve noticed that most people lie about their opinions when they’re talking to stupid people about their intelligence.

  16. Maybe Kahneman’s book is more nuanced, but Lehrer makes it all sound incredibly stupid. He ‘proves’ that smart people are more prone to make knee-jerk mistakes … and somehow extrapolates from that a vast problem with the act of introspection.

    But even if the ‘proof’ is true (and it may be), he doesn’t for a second give consideration to what happens in the person’s mind in the AFTERMATH of the initial judgment. Stupidity (and insanity) is defined by its inability to do anything different. Smart people MAY have tangled-up biases, but they’re more prone to reconsider and correct themselves, especially when ambushed with trick questions about trees and baseballs.

    1. The book is more nuanced but is rather tediously aimed at the business management crowd.    I made it about 3/4 through and skimmed the rest.
      Parts of it were very interesting.

    1. That’s not really true. It’s administered by the Nobel Foundation, the laureates are chosen by the Royal Swedish Academy of Sciences like the other science prizes and it’s presented at the same ceremony, so there’s nothing to distinguish it from other official Nobel Prizes except that it wasn’t in the original set.

      1. Which is the the only distinction that counts.  Nobel endowed five prizes.  Those are the Nobel prizes. 

        Everything else may be as  or even more prestigious than the Nobel prizes, but they aren’t Nobel prizes.

          1. No, I just think that, of all the possible criteria that you could have selected, you’ve chosen the least relevant one.

    2. The absence of a nobel prize for a field means little if there ain’t even one for mathematics. And we all know the story behind that one.

  17. Everyone is stupid. Including you. (And me, of course.)

    I have found this to be a very healthy mantra to keep me grounded in reality.

  18. When it comes to making decisions on behalf of/affecting many other people:

    empathy, honesty, humility >> intelligence, insight, expertise

    The latter are good for designing systems and tools, but not for wielding those designs, especially in high-stakes situations.

  19. I’m paid to help a Chinese company create new products.  It’s full of bright people who are great at copying other products.  But they struggle at coming up with something new.  It’s a struggle to get them to stop saying “Company X does this.”

    It was an epiphany when I figured out what they problem is.  They don’t know how to fail.  Making a mistake is a foreign concept.  So I’m spending quite a bit of time showing them how to go out on a limb no one else has, fall off, pick themselves up, and do it again.  And again.

    You would think going out on a limb would be the hardest part.  Nope.  It’s realizing that the limb broke and they are flat on their back.

    The only thing I can think is it must be a byproduct of the education system.  Every problem has a textbook solution.  Never question the solution, because that would mean the textbook was wrong.  Nevermind they grabbed the wrong textbook and misapplied it.

    Bright people making dumb mistakes and never questioning it.

  20. I dont know how many others actually have this book but I am currently reading it. Its very good and I think the conclusions some folks have made here are too simplistic. Read it then you can judge. Its a perfect example of those who make snap judgements without having all the information easily available to them.

  21. Well, I don’t know if it makes me smart or stupid, but this book was one of the most boring and badly written books I’ve read in years. Basically a shopping list of dozens of different studies, and little in the way of significant revelation bubbling up from them all. Tried twice but could not manage to finish it…..

  22. That’s what I like about Dunning – Kruger: It’s FAIR.

    So fair that it makes room for EVERYBODY.

  23. The “best and the brightest” who worked for LBJ and the neocons who worked for GWB both managed to launch and wage catastrophic wars, in large part because of blindness of the history of the peoples and the regions in which they were fought.
    As for SATs, I always thought they only measured how good a guesser you are.

  24. Working on a beach profile for an introductory oceanography class, I learned an important lesson. I made a simple mathematical error and the professor chuckled and pointed it out to me. “Your calculations show the water was fourteen feet deep. But you were standing up to your chest holding the stadia stick. How tall ARE you?” 

    The take home lesson? Always double check your answers to be sure they make sense!

  25. What bothers me about the recent interest in cognitive biases is how I’ve seen some people use it to simply wave away rationality and discourse – to say, in effect, “people like you aren’t as smart as you think you are, so you should shut up.  The end.”  You could easily apply such an argument to, say, climate change, or anything that people want to actually argue about.

    The thing is that discourse is obviously the antidote to cognitive bias.  If the  rationale behind an argument is faulty, you have the power to get in there and show it.  To throw up your hands and say “I don’t have to listen to you because cognitive bias” is meaningless.

  26. I believe the most important discoveries were “stupid mistakes” rather than planned calculated, so it’s all good. Besides smart / stupid what is that? How do we measure it? I agree with this delightful BBC documentary on stupidity, this is no simple question to answer. 

  27. To me it seems that the conclusion can be applied to about any human ability. If we are sure of our abilities we tend to rely on them instead of on the tools which can prevent errors and accidents. As these tools have a long history and a lot of (tried) thinking behind, their use can actually make you smarter.

    Persons that regard themselves as physically strong and have not had a real accident are much more likely than the “weaklings” to try to do things with their own hands and injure themselves. The weaklings might have after all an incentive to find the right tools or to divide up the job in a reasonable manner, that is to think about what exactly they are doing and how before they actually do it.

    There are tools to aid human reasoning and there is knowledge of human reasoning and it’s possible failings going back thousands of years, now strengthened by empirical science. No matter how smart you are, you forgo them at your own risk.

  28. I haven’t read the book, I have read the New Yorker piece.  It seemed to discount the ability of (at least some) smart people to listen to feedback and consider and correct their own biases.  Others above have made the same observation / response.

    I agree with the “This argument can be used to say ‘I don’t have to listen to you because of cognitive bias'” comments above as well.

    The hardest part is balancing being aware of cognitive biases and blind spots and information gaps, and still being aware and decisive enough to be very effective (at work, in life, etc).  Plenty of people use this all as an excuse to either just block out awareness of their own limitations or become incredibly indecisive.  Neither of those works.

  29. Two questions:

    1) What are the cognitive biases of psychologists investigating cognitive biases?

    2) Does Diesel have anything to do with this?

  30. Was surprised when I saw Lawrence Lessig reference this article.  To talk about intelligence – “smart people” – without taking into account personality types doesn’t make sense to me.  For example, I’m an INTJ personality type – best characterized by the saying “know what you don’t know”.  I tend to question my perspective more than I question others.  I’d guess that there are many smart people who are also INTJ personality types.

Comments are closed.