Seth Roberts' Final Column: Butter Makes Me Smarter

Mr. Roberts ate half a stick of butter ever day for a year.

(Wikimedia Commons)
(Wikimedia Commons)

Note: Seth Roberts submitted this column to Betabeat before his untimely death. We publish it now with a heavy heart and per his request will be making a donation to Amnesty International.

A few weeks ago my sister sent me a link to an article (“Butter is Back”) by Mark Bittman, the New York Times food writer. I told her I’d clicked on a link to the article but had forgotten to read it. She was incredulous. How could you not want to say “I told you so”?

Here’s what Bittman said:

A meta-analysis published in the  journal Annals of Internal Medicine found that there’s just no evidence to support the notion that saturated fat increases the risk of heart disease. (In fact, there’s some evidence that a lack of saturated fat may be damaging.)

In a 2010 talk I had said the evidence against saturated fat was weak. If you were there, I told you so.

I found something else Bittman said more interesting:

The real villains in our diet — sugar and ultra-processed foods — are becoming increasingly apparent.

I see. The experts were utterly wrong, for decades, about saturated fat . . . but they couldn’t possibly be wrong about “sugar and ultra-processed foods”.  What’s increasingly apparent to me is that nutrition experts shouldn’t be trusted.

My doubts are long-standing. In 1990, I found breakfast made me wake up too early. Experts said breakfast is the most important meal of the day. In 2000, I found that drinking unflavored sugar water made me lose weight, which led to The Shangri-La Diet (2006). Every expert said sugar is fattening.

The general rule behind the Shangri-La Diet is that calories without smell cause weight loss. Unflavored sugar water has no smell. After the book appeared, I was asked if the calories could come from flaxseed oil, which in capsules has no smell.

That was an interesting idea, so one evening I swallowed five flaxseed oil capsules. The next morning, I put on my shoes standing up, which meant I tied my shoelaces standing on one foot. I’d done this hundreds of times. That morning it was much easier.

Had flaxseed oil improved my balance? It wasn’t a crazy idea. The brain has lots of omega-3 and flaxseed oil is high in omega-3. (That experts said flaxseed oil is a poor source of the omega-3 in the brain didn’t worry me.)

I devised a test of balance and carefully measured the effect of flaxseed oil. It did improve my balance. It also improved my performance on other brain tests. I tried different doses. The best dose – the smallest dose that gave me the best balance — turned out to be 3 tablespoons/day, which was much more than flaxseed oil bottles recommended (1 tablespoon/day). Soon after this, my gums went from red (inflamed) to pink (not inflamed), no doubt because omega-3 is anti-inflammatory. Several readers of my blog found the same thing.

The clarity of the improvement stunned me. Professional scientists had found that omega-3 helped children with attention deficit hyperactivity disorder (ADHD) and adults with bipolar disorder, but those two groups are a small fraction of the population and the experiments had been hard and slow. My experiments had been easy and fast. I did not have bipolar disorder. I ate fish several times per week. The latest Dietary Guidelines for Americans said if you eat that much fish you will get enough omega-3. My results implied that was wrong, at least for me.

More broadly, the results suggested ignorance, that we know little about how food affects the brain. It was no surprise we need omega-3, yet even here it seemed the experts had gotten the wrong answer to the question of how much. The results also suggested I might be able to fill in some of the blanks, at least for myself.

To learn more about the effect of food and other things on my brain, I started to track my brain function. I used a simple reaction-time test. The test consisted of several easy arithmetic problems (e.g., “3+5”). I saw each problem on my laptop screen and typed the answer as fast as possible. I did the test daily. It detected the effect of flaxseed oil. Maybe it could find other things that mattered.

In 2009, I bought some pork from Devils Gulch Ranch, a family farm in California. I ate the familiar cuts and was left with an unfamiliar cut that was almost all fat (which I later learned was pork belly). At the time, like all sensible Americans, I avoided animal fat. The pork belly repulsed me but I didn’t want to throw it out.

It stayed in my freezer for months. Finally I ate it. The next morning I woke up full of energy, much more than usual. It must have been the pork belly. I felt full of energy all day. Eventually I did experiments to measure the effect of pork belly on my sleep. I woke up more rested after days I ate pork belly.

These results made me stop believing saturated fat is bad. The results were very clear. Sleep is central to health. Something that improves your sleep is likely to improve your health. My experiments were far from perfect but I found them more convincing than any of the evidence used to claim saturated fat is bad. Almost all that evidence came from surveys that compared people who ate more saturated fat with people who ate less. The two groups always differed in other ways (e.g., income) besides fat intake.

I tried to eat pork belly every day. One day it was going to be too hard to do this, so I ate butter instead. At lunch, at a restaurant, I asked for more butter twice. A few hours after lunch, I felt a pleasant warmth in my head. Pork belly hadn’t done that. Maybe butter was better for the brain than pork belly. I switched from pork belly to butter.

A few days later, my arithmetic scores suddenly improved. This graph shows my daily scores.

The upper graph shows all my scores up to the day with the outlier (the rightmost day). The day with the outlier was unusual in several ways. For example, I ate a lot of butter; I also stood on a Chinese cobblestone mat. The next day I repeated all the unusual features and got an unusually fast score again. That’s what the lower graph shows.

After that – after I was sure the improvement was repeatable — I tested each unusual feature separately. It turned out that the one that mattered was eating butter. So I continued to eat butter daily – half a stick (about 60 g) per day. My arithmetic speed remained faster than usual, as this graph shows.

After I gave a talk about this, a cardiologist in the audience said I was killing myself. It turned out he didn’t understand the evidence that animal fat is bad. He thought the Framingham study found an association between animal fat and heart disease. In fact, it found no difference in diet between those who had heart attacks and those who didn’t (so the name was changed from the Framingham Diet Study to the Framingham Heart Study).

Was I unusual? My talk inspired Greg Biggers and Eri Gentry of Genomera, a Palo Alto company, to run a test with about 40 subjects to see if butter had the same effect on other people. The subjects were randomly assigned to one of three conditions: butter, coconut oil, or nothing. All subjects did daily arithmetic tests. Subjects in the butter condition got faster; subjects in the other two conditions didn’t change. Apparently I wasn’t unusual. That coconut oil didn’t produce improvement argues against a placebo-effect explanation. (Nor does a placebo-effect explanation explain the first outlier – the sudden improvement.)

I wondered about the best amount of butter – half a stick seemed like a lot.  So I tried a smaller amount. For eight days I ate 30 g/day butter; before and after that I ate 60 g/day. Here is what happened.

I was slower with 30 g/day than with 60 g/day. I had no reason to think 60 g/day would be better than 30 g/day – in fact, I hoped 30 g/day was better. So these results also argue against a placebo-effect explanation.

It was nice to know all that but I did wonder: Was I killing myself? Fortunately I could find out. A few months before my butter discovery, I had gotten a “heart scan” – a tomographic x-ray of my circulatory system. These scans are summarized by an Agatston score, a measure of calcification. Your Agatston score is the best predictor of whether you will have a heart attack in the next few years. After a year of eating a half stick of butter every day, I got a second heart scan. Remarkably, my Agatston score had improved (= less calcification), which is rare. Apparently my risk of a heart attack had gone down.

Seth Roberts was a professor of psychology at Tsinghua University (Beijing) and author of The Shangri-La Diet. He blogged at Seth’s Blog.

Seth Roberts' Final Column: Butter Makes Me Smarter