Seth’s blog

• A skeptical look at Karl Popper by Martin Gardner
• Effects of food on the brain
• fat sensitivity correlates with BMI (less sensitive, greater BMI)
• Bill McKibben on Climategate: “If you managed to hack 3,000 emails from some scientist’s account, you might well find a few that showed them behaving badly, or at least talking about doing so.” This gives new meaning to the term denialist.
• if someone steals $2 million do you still trust their research?

Thanks to Vic Sarjoo, Anne Weiss, and Marian Lizzi.

Yesterday I blogged about a sudden improvement in how fast I could do arithmetic. The improvement was much larger than normal variation and happened after I did four things that I rarely did. In chronological order:

1. Ate about 30 g of butter.

2. Stood on a cobblestone mat (for 5 minutes, which was all I could bear).

3. Stood during the test.

4. Walked for 10 minutes just before the test.

To find out which mattered, I did them again in the same order and at the same times of day, but with tests before and after each one.  If performance suddenly improved after one of them, then I’d know.

Here’s what actually happened.

The last six points are the relevant results. The first of the six points (627 msec) was before everything. The second (613 msec) was after butter but before the cobblestones. The third (630 msec) was after the cobblestones but before standing. The fourth (610 msec) and fifth (603 msec) were while standing but before walking. The final one (581 msec) was while standing after walking.

I was surprised and pleased how closely the first and last scores repeated the earlier difference. The first score was close to the previous baseline; the last score was close to the previous outlier. A big improvement seems to be under my control.

Before doing these tests, my best guess about what caused the improvement was the walking. But the scores were improving before the walking so that’s unlikely. Perhaps the walking was one of several factors that helped. The data suggest, if anything, a shocking conclusion: butter made my brain work better. An alternative, less consistent with Occam’s razor, is that butter, standing, and walking all produced smaller improvements, which together added up to the big improvement. The cobblestones produced a short-lived decrement.

That pork fat improved my sleep obviously supports the butter interpretation. I should be less surprised than anyone else, but still . . . Last week I noticed something else that supports the butter explanation. At a restaurant with a friend, the waiter brought bread and olive oil. I asked for butter. I spread all of it on a piece of bread, then asked for more butter, and spread all of that on another piece of bread. (About 30 g butter total.) It was the first time I’d eaten a large amount of butter at a meal. An hour or so later, I felt unusually good, some combination of calm and warmth. I never noticed this after eating pork fat, but butter may be to pork fat as hamburger is to steak: Easier to digest. The pork fat is within cell walls; the butter fat isn’t.

If you read Good Calories Bad Calories you may remember that the Arctic anthropologist Vilhjamur Steffanson spent a year on an all-meat diet, with no ill effects. (In an earlier post about Steffanson, I stressed the fermented food that the Eskimos ate.) You may not know that animal fat was crucial for his health during that year, which began with a brief attempt to eat lean meat (meaning meat without fat):

On February 26, 1968, [Stefansson] was admitted to the ward and on February 28, started on the meat diet. At our request he began eating lean meat only, although he had previously noted, in the North, that very lean meat sometimes produced digestive disturbances. On the 3rd day nausea and diarrhea developed. When fat meat was added to the diet, a full recovery was made in 2 days.

During the year, he got about 80% of his calories from fat.

Via Inhuman Experiment.

Kathy Tucker draws my attention to a recent article about the ketogenic diet, which is essentially a very-high-animal-fat diet, used to treat childhood epilepsy. I’ve blogged about the ketogenic diet (here, here, and here) but that was before I was on a similar diet. Kids on the diet didn’t develop high cholesterol (”very few children actually end up with cholesterol or lipid problems on the diet”). I slept better when I ate more animal fat, which suggests that animal fat makes the brain work better overall. The success of the ketogenic diet supports that idea. My results suggest that it is the animal fat, not the other fat, that makes the diet effective.

That many kids with epilepsy get better when put on the ketogenic diet can be seen as a canary-in-the-coal-mine phenomenon. Canaries are more sensitive to bad air than miners; children with ketogenic-responsive epilepsy are more sensitive to lack of animal fat than the rest of us. That lesson was lost on me when I first learned about the diet and its success. The broader lesson is that almost any disease has something to teach us about what the best environment is.

After learning that animal fat improved my sleep, I happily ate much more of it. I wasn’t worried that it made something else worse (e.g., heart disease). I believe that all parts of our bodies have been shaped by evolution to work well on the same diet, just as all electric appliances are designed to work well on the same house current.

A to-be-published meta-analysis in the American Journal of Clinical Nutrition supports my view that animal fat is nowhere as bad as we’ve been told a thousand times. It says:

During 5–23 y of follow-up of 347,747 subjects, . . . intake of [more] saturated fat was not associated with an increased risk of CHD [coronary heart disease], stroke, or CVD [cardiovascular disease]. The pooled relative risk estimates that compared extreme quantiles of saturated fat intake were 1.07 (95% CI: 0.96, 1.19; P = 0.22) for CHD, 0.81 (95% CI: 0.62, 1.05; P = 0.11) for stroke, and 1.00 (95% CI: 0.89, 1.11; P = 0.95) for CVD.

Emphasis added. One aspect of the results suggested that studies that found an positive association (more fat, more disease) were more likely to be published than those that didn’t find an association or found a negative association. Which means these numbers may underestimate the good effects.

Thanks to Steve Hansen and Michael Pope.