Will eating a half a stick of butter a day make you smarter?

Does eating a lot of butter every day improve brain function? To find out, Eri Gentry (co-founder of BioCurious and a MAKE contributor) is calling for volunteers to eat three tablespoons of butter or coconut fat each day and take a daily math quiz.
201010140957 Will eating one of these fats improve your math performance?  Based on Seth Roberts' butter and math study, recently presented at a Bay Area Quantified Self Show & Tell, during which Seth ate half a stick of butter each day and performed better in math, we expect the answer to be yes.  

To test this theory across multiple participants and two types of fats, from October 20 - November 12, I will run Butter Mind & Coconut Mind, meant to collect data on the ingestion of specific "food fats" and math scores/times.

To qualify for the study, you must be willing to eat 3TBS of butter or coconut oil (sticking to the same one) for 8 days and do a 32 problem simple math test for 24 days.  You must have access to the internet to submit your scores.

Will Butter Make You Smarter? Introducing Butter Mind...and Coconut Mind



  1. Experiment needs a control. You need people to take the math test without eating either fat. I’ll volunteer for that.

    You also need people to track their butter/coconut fat intakes that may come from other dietary sources.

    1. You also need the people to experience the same exact lived conditions for the experiment duration…which is impossible.

      1. You’re right, the normal daily fat intake needs to be quantified for each of the test-takers. While this is not feasible, it is to do a dietary survey and attempt to standardize groups based on that. That’s better than ignoring it completely.

        But, as is pointed out several times already, the results are meaningless without an additional control group anyway.

        I’m pretty disappointed. I would have expected the founder of Make and BioCurious to be a little more familiar with basic science principles than this.

      2. @Louis A., no you don’t. With enough subjects in the experiment, intervening effects along the lines of what you describe would be randomized out.

    2. For the control substance, I nominate “I Can’t Believe It’s Not Butter”, if they still make the stuff.

  2. Ever since I went paleo and stopped eating as much pasta (and consequently butter) I have noticed a decrease in my math skills.

  3. I’d say you can’t draw any conclusions from this experiment. You would at least need a control group that doesn’t eat “fat” but does take math tests for 24 days, or group that (unknowingly) takes placebos and does the tests. Still interesting, though.

  4. I did this once; sort of. I was on a cyclical Ketogenic diet or CKD. on the days where I had a high fat intake I would notice that I was able to think better and had a better mood. My grades in school improved as well. Not sure if this is from the fat or the ketones acting as anti-inflammatories, or BHB.

  5. Needs a control and a double-blind too. You need to have some people eating a really good “I can’t believe it’s margarine!” margarine.

  6. Why not? If the final stages of syphilis can turn you into a genius before you die I’m sure other unhealthy things can make you smarter as well.

  7. This is terrible. If the participants believe that better might make them smarter, it is very likely that their tests will show the participants improving. There have been many, many studies that show that people perform differently on tests if they expect to do better.

    Not to mention the effects of practice… Oh boy, do you think that taking a math test on the same subject every day for three weeks might cause them to improve their scores? That’s just dumb design.

    As others have said, this requires a bunch of modifications to make it even slightly sound:

    1) Don’t tell people what it is you expect to see

    2) Have a control that eats something every day that is not butter or a fat (it’s no good having a control group that knows that they’re a control — that’s as bad as no control at all)

    3) Don’t use self-selected participants (“oooh, ohhh, I love butter and math tests!”)

  8. Ah, so 3 tablespoons of butter = one-half a “stick”.
    Good. Those are 50 ml. tablespoons, right?
    So now I know how to convert those recipes that call for “sticks” of butter into something I understand, like tablespoons. Thank you!

    1. When you buy sticks of butter, it is usually labeled on the wrapping. 8 tablespoons to a stick (not 6, as the article would seem to indicate).

      I never knew this until recently. Growing up, my family bought pounds of block butter that wasn’t subdivided into sticks.

  9. If you’re stupid enough to sign up for this thinking it’s a legitimately structured scientific study, then you’ve got nowhere to go but up, so it may just work.

    1. Anon #24: thanks for the input – my butter comes in 1-lb blocks like in the photo above.

      On Topic: Buttered brains? Hallowe’en must be just a fortnight or so away, then.

  10. Your brain is mostly fat. Nutrients are carried in to the brain by fat. I think a lot of contemporary cultural stupidity can be blamed on low-fat and fat free foods, ie peoples brains not getting sufficient nutrition for clear thinking/emotional stability.

    1. Your body is fully capable of interconverting fats and sugars/carbohydrates. If it couldn’t, then eating sugar couldn’t make you gain weight, and you’d starve if you ate a low-carb diet, neither of which is what actually happens. So, is there any evidence that eating too little fat, or is that just speculation?

      1. So, is there any evidence that eating too little fat, or is that just speculation?

        Since some nutrients are fat soluble, the need to eat some amount of fats would seem to be a given.

  11. eating half a stick of butter a day is not smart, maybe after eating it for a few weeks they’ll be able to figure that out.

    don’t get me wrong, i love both butter and math (not together)…but this is just dumb.

  12. It’s a TRICK! The REAL test is to see if eating a half stick of butter every day will make you fatter, but nobody would sign up for that.

  13. In Borgotavia everybody knows that fat helps intellect, because the brain is made of fat and uses a lot of fat. In Borgotavia, however, it is considered dumb to eat fat for brains, and they eat directly the brains.

  14. If you make your Mac N Cheese with butter, and perhaps many *in college* do, could it be a brain boost?

  15. Since both coconut oil and butter are pretty high in saturated fat, which is needed for brain development, both should increase brain function.

    To those trotting out the old “hurr hurr, this will make them fatter”, I drastically increased the amount of (specifically saturated) fat in my diet and lost a whole bunch of weight, so it doesn’t happen.

    Simplifying greatly:
    Eat sugar/starch, burn sugar/starch, store excess as fat thanks to insulin.
    Eat no sugar/starch, burn fat, use those stores of excess fat.

  16. It seems to me that just talking about eating butter is making us all smarter, so actually eating it – doubly so! *nom*

  17. Weston Price Foundation recommends butter every morning (REAL butter, preferably raw butter if you can find it). I can’t say that it has made me smarter, but I feel a whole lot better. My energy and attention are much more balanced between 8am and my lunch at 1pm. When I eat all carbs & coffee at 8am, it’s a rollercoaster and I’m looking for food before noon.

    1. Oops, I only read the linked page after they added the control group. Sorry for snarking at those of you who commented before it was edited.

    2. Bwraagh! It’s still bo good, even with the silly “control” group added. It’s not a control if the participants know that they’re the control group!

      Scientist: I believe that butter makes you smart! In this test, I will give you butter. Let’s see how you do.
      Participant 1: Ooh, like many, many studies have confirmed, if I think I’m going to do better I will! A=5! x^2 + 3! E=mc^2!
      Scientist: I believe that butter makes you smart, and for you, participant 2: no butter.
      Participant 2: Awww, I suck at everything, and now I don’t even get to be in the smart kid’s group. No butter for me, and so I will do terribly in the tests.

      This =/= science

  18. For everybody saying this needs a control, did you read the link? It clearly says that all participants will be randomly assigned to one of three groups: a butter group, a coconut oil group, and a control group that does not eat additional fats. Can we do a study on whether butter improves reading comprehension next?

  19. Fat does not cross the blood brain barrier. Sugar is the only nutrient that crosses the blood brain barrier. If fat has an impact on cognition, it must therefore be indirect.

  20. The people who say *this* run of the experiment needs to have a control group “in order to reach any valid conclusions” are wrong and have basically missed the entire point of Seth’s approach to science. You started out asking two really good, really simple questions:

    (1) Are Seth’s results with regard to a butter consumption/math performance relationship reproducible for other people besides Seth?

    (2) Does coconut oil produce similar results under similar circumstances?

    Your original study design sufficed to answer THOSE SPECIFIC QUESTIONS. If you can find 2 people or 5 people or 10 people or however many who are willing to eat lots of fat and take lots of math tests, you would probably do best to put *all* of them towards answering those two questions and not waste *any* on anything else.

    It is perfectly plausible that when you do the study you’ll find the answer is NO – Seth’s results aren’t reproducible using this test design. Maybe your test will find no relationship at all. Or maybe some people will show the exact reverse relationship – they do worse with more butter instead of better. Or perhaps some distinct demographic group or behavioral group seems to do better or worse with more fat – just males or just old people or just people in a certain weight range – and for other groups it doesn’t matter so much. If you waste half your test subjects in a “control” group that consumes no fat, you’re much less likely to answer the exact question you were asking – is this effect generalizable?

    IF the simple test turns out to suggest that, yes, there really is a measurable effect for some subset of people following some specific test protocol, THEN you can design more elaborate tests to tease out WHY there is an effect and how strong it is. Including varying the test protocol to establish a dose-response curve and rule out all sorts of alternate testable hypotheses that might explain the effect. But trying to design a rigorous foolproof study to anticipate all possible objections to an effect you haven’t yet even verified *exists at all* is a waste of time.

  21. this is an odd experiment how could eating butter help you with math.
    i dont think it is possible i believe that it could actually just harm
    the body because of all the saturated fat that the person will consume.
    there are other methods of getting better at math like studying instead of
    eating butter life isnt that easy why are people always trying to find the
    easy way out if math is hard for a person then they should study if they
    want to get better at math thats how it has been before why is butter going to change that.

  22. As much as I’d like an excuse to slather butter on everything, I can’t help but think that *anyone* doing math (or any mental test) regularly will naturally improve, diet notwithstanding.

  23. 3 tablespoons would be a tiny fraction of the fat I eat each day…

    I wouldnt be surprised to find an association between good-fat consumption and intelligence, but I’ve noticed plenty of intelligent people reject government low-fat advice that is based on poor science… so intelligence may be driving people to eat butter…

  24. The math quiz is incorrect. If you watch the talk Roberts gave you will see that he was not testing mathematic ability, he was testing cognitive response time. He did not have to solve math problems he had to simply recall the answer to a math problem. A better experiment would be to do the game concentration and see how many answers were right. If there is an improvement then the butter may have a correlation. As Seth said in the talk, rarely will you run stats tests and find out that you are proving yourself wrong about something, typically it is proven right and this experiment led to an accidental realization about the link between butter and memory.

Comments are closed.