Summary: Rigorous scientific experiments are hard to apply in daily life but we still want to try out and evaluate things like self-improvement methods. In doing so we can look for things such as a) effect sizes that are so large that they don't seem likely to be attributable to bias, b) a deep understanding of the mechanism of a technique, c) simple non-rigorous tests.

Hello there! This is my first attempt at a top-level post and I'll start it off with a little story.

Five years ago, in a kitchen in London...

My wife: We're going to have my friends over for dinner and we're making that pasta sauce everyone likes. I'm going to need you to cut some red peppers.

Me: Can do! *chop chop chop*

My wife: Hey, Mr. Engineer, you've got seeds all over! What are you doing to that pepper?

Me: Well, admittedly this time I was a bit clumsy and there's more seed spillage than usual - but it's precisely to avoid spilling seeds that I start by surgically removing the core and then...

My wife: Stop, just stop. That's got to be the worst possible way to do this. See, this is how you cut a pepper, *chop chop chop*. Nice slices, no mess.

Me: *is humiliated* *learns*

Now, ever since then I've cut peppers using the method my wife showed me. It's a much better way to do it. But wait! How do I know that? Don't I realize that humans are subject to massive cognitive biases? Maybe I just want to please my wife by doing things her way so I've convinced myself her method is better. Maybe I'm remembering the hits and forgetting the misses - maybe I've forgotten all the times my method worked out great and the times her method failed. Maybe I am indeed making less of a mess than I used to but it's simply due to my knife skills improving - and that would have happened even if I'd stuck with the old method. And there are probably a dozen more biases and confounding factors that could come into play but I haven't even thought of.

Don't I need to do a test? How about cutting up lots of peppers using the two alternative methods and measuring seed spillage? But, no, that's not good enough - I could subconsciously affect the result by applying less skill when using one method. I'd need a neutral party to operate the methods, preferably a number of people. And I'd need a neutral observer too. The person who measures the seed spillage from each operation should not know which method was used. Yeah, a double blind test, that's the ticket. That's what I should do, right?

No, obviously that's not what I should do. There are two reasons:

A) The resources needed to conduct the suggested test are completely disproportional to any benefit such a test might yield.

B) I already bloody well know that my wife's method is better.

The first reason is obvious enough but the second reason needs a bit more exploration. Why do I know this? I think there are two reasons.

* The effect size is large and sustained. Previously, I used to make a mess just about every time. After I switched methods I get a clean cut just about every time.

* I understand the mechanism explaining the effect very well. I can see what's wrong with the method I was using previously (if I try to pull the core through a hole that's too small for its widest part then some seeds will rub off) and I can see how my wife's method doesn't have that problem (no pulling the core through a hole, just cut around it).

I'd like to try to generalize from this example. Many people on this site are interested in methods for self-improvement, e.g. methods for fighting akrasia or developing social skills. Very often, those methods have not been tested scientifically and we do not ourselves have the resources to conduct such tests. Even in cases where there have been scientific experiments we cannot be confident in applying the results to ourselves. Even if a psychology experiment shows that a certain way of doing things has a statistically significant1 effect on some group that is no guarantee that it will have an effect on a particular individual. So, it is no surprise that discussion of self-improvement methods is frequently met with skepticism around here. And that's largely healthy.

But how can we tell whether a self-improvement method is worth trying out? And if we do try it, how can we tell if it's working for us? One thing we can do, like in the pepper example, is to look for large effects and plausible mechanisms. Biases and other confounding factors make it hard for us to tell the difference between a small negative effect, no effect and a small positive effect. But we still have a decent chance of correctly telling the difference between no effect and a large effect.

Another thing we can do is to use some science. Just because a rigorous double blind test with a hundred participants isn't practical doesn't mean we can't do any tests at all. A person trying out a new diet will weigh themselves every day. And if you're testing out a self-improvement technique then you can try to find some metric that will give you an idea of you how well you are doing. Trying out a method for getting more work done on your dissertation? Maybe you should measure your daily word count, it's not perfect but it's something. As xkcd's Zombie Feynman would have it, "Ideas are tested by experiment, that is the core of science."

Erring on the side of too much credulity is bad and erring on the side of too much skepticism is also bad. Both prevent us from becoming stronger.

1) As good Bayesians we, of course, find psychologists' obsession with null hypotheses and statistical significance to be misguided and counterproductive. But that's a story for another time.

New Comment
68 comments, sorted by Click to highlight new comments since:

See also Seth Roberts, who's done a lot with self-experimentation by way of many small low-cost experiments looking for large effects.

Seconded.

Reading through his blog shocked me at just how much important stuff you can learn by simple tests.

I've personally experienced several benefits from the results of his/my self experiments. The most noteworthy to this community would be that I have eliminated 3 different causes of mental fatigue. For instance, due to changes in diet, I can go without food all day and still be mentally productive, whereas before, if I hadn't eaten in a few hours, I wouldn't be able to think clearly enough to be productive.

[-]pjeby120

For instance, due to changes in diet, I can go without food all day and still be mentally productive, whereas before, if I hadn't eaten in a few hours, I wouldn't be able to think clearly enough to be productive.

Details? Please? ;-)

In short, replace carbohydrates with animal fats.

I still get hungry if I don't eat all day, and it can still have some negative impact, but it's an order of magnitude less of a problem.

It took a little while for my body to adapt to the changes. I didn't actually take notes, and it was sort of a gradual transition for me, so I'm not sure how long everything took. I'd expect a lot of the benefits to come after the first meal, but by a month or so I think most things stabilized and my tastes adjusted to fit my diet better.

If anyone wants to know more details, I'm more than happy to share, but I'm not sure exactly which parts would be interesting. Just PM me and we can chat.

Why animal fats in particular? This is somewhat disheartening as I find both the idea and the taste of eating mammals to be physically nauseating. Poultry and fish aren't as significant for me, but I don't eat much of those either.

Since you've significantly increased your animal fat intake, how has your animal protein intake changed? Do you just add lard to your diet, or do you eat really fatty cuts of meat?

Partly for the saturated fat, partly for less extreme amounts of omega-6 (though some plant oils are not as bad as others), and partly the personal fact that my stomach doesn't seem to like significant quantities of olive oil, but craves large amounts of animal fat. Whipping cream and butter are both mammal fat that don't require killing the animal - eggs fried in butter are quite tasty :)

I never really monitored the amount of protein I eat, but it has to have gone up. I do eat fatty cuts of meat, but I add a lot of nearly pure fat to my diet, so it's a bit of both.

Good point.

I like butter, though not as much as olive oil, but I also eat a good deal of cheese and eggs, so I guess I already eat a fair quantity of animal fat, just not animal body fat.

Good Calories, Bad Calories is a comprehensive and detailed discussion of why the conventional wisdom on diet is flawed. Here's a less rigorous run down of the different fats. Saturated fat is the main reason to prefer animal fats as it is hard to find in non animal sources. One non animal derived option is coconut oil.

Most people who question conventional wisdom are cranks. As a non-expert, I am not capable of evaluating the object-level arguments on the subject. Why should I grant Gary Taubes any more credibility than any other random person who writes a book with lots of footnotes?

[-][anonymous]30

I haven't read the book either.

I don't know about heart disease and cancer risk and I'm in roughly the same epistemic position as you on that.

For plain old fat loss, though, there's more reason to believe that more fat and protein compared to carbohydrates is effective. First, that's what athletes and their nutrition coaches do -- it's always useful to look at the practices of people whose livelihood depends on objective success. Second, I've read a fair number of CDC studies showing that diets have modest to zero effect -- and usually the test diets do not have a high protein + fat to carb ratio. That's weak evidence, but evidence: the stuff that doesn't work is not the candidate in question. Third, it just makes sense that simple sugars break down into glucose faster than anything else, which means a quick spike in blood sugar instead of slow release over time, which means you'll be hungry soon after. No, plausible mechanisms aren't science, but they help. Fourth, it's easy enough to perform the experiment yourself for three months and see what happens.

If you are not confident in your own ability to recognize cranks (either in general or in a particular field) then sticking with conventional wisdom may be your best option. In general my experience is that it is not very difficult to spot genuine cranks however but YMMV. I'd suggest that health in general is a poor place to put great faith in conventional wisdom however since there are many powerful interest groups influencing things like government issued recommendations whose interests are not always well aligned with yours.

Nutrition and diet are relatively amenable to self experimentation however and so at least some claims are fairly easily testable. Obviously you aren't going to be able to confirm claims of reducing your risk of heart disease or various cancers (for example) in the short term but claims about weight loss are trivial to verify and it is not terribly difficult to test other claims by having blood work done (triglyceride levels, LDL/HDL ratios etc.) and by keeping a record of possibly diet related symptoms under different regimes (mood, energy level crashes, digestive problems, etc.).

Thanks for the information. Although I haven't read it, I'm familiar with the topics covered by Good Calories, Bad Calories, and feel that the points it brings up have good support among people who care about the scientific method.

The two pages on Mark Sisson's website that you linked to only claimed that saturated fat isn't bad; I can agree with that, but what I'm really interested in is why you seem to feel that they are especially healthful. It doesn't seem like he claims that.

but what I'm really interested in is why you seem to feel that they are especially healthful. It doesn't seem like he claims that.

Well he tends to emphasize that the key to healthy eating is to minimize refined carbohydrate intake and he recommends a protein intake in the range of 0.7-1.0g per lb of lean body mass so fat intake is in part driven by the need to make up the rest of your target calorie intake. Saturated fat seems to be the a fairly healthy way to make up the calorie deficit given the options.

That a much less dramatic claim than it sounded like you were making at first, which is good for me, since it doesn't make me feel obliged to add animal fat to my diet.

I do eat a lot of fat, though mostly unsaturated (nuts and avocado). I also eat a lot of carbs, but I try to eat almost no refined carbohydrates; your example of beans and rice amuses me because brown rice and beans are a significant portion of my diet.

your example of beans and rice amuses me because brown rice and beans are a significant portion of my diet.

I think beans and rice was actually jimmy's example but yeah, rice is pretty horrendous health wise, brown or not. The jury's still out on beans from what I've read. You can't easily match the calorie intake from grains without upping your fat intake though since you need to eat a hell of a lot of green vegetables to match the caloric value of grains.

Sorry about the confusion.

I've been told by dozens of people that rice is "pretty horrendous" health-wise, and I've looked into it a bit myself. The signal-to-noise ratio is really low on this issue, so I keep running into very unscientific claims. The closest I've come to good evidence for that proposition is weak but scientific evidence that glucose intake (as part of starch) needs to be balanced with fiber intake, and that foods like rice must be balanced with foods that are almost entirely fiber (and have virtually no calories).

Would you please link me to more good arguments against consuming all starches, not just refined carbohydrates?

[EDIT: replaced fructose with glucose]

A couple of Mark Sisson's articles on grains. I'm not linking to this site because it is the most rigorous or most scientifically unassailable source for nutrition information but simply because it's easy for me to find reasonable summaries there. Good Calories, Bad Calories would be my recommendation if you want exhaustive references to the studies that many of these claims derive from. Googling for 'paleo diet' or 'paleolithic diet' should turn up a bunch more information as well.

I don't really take Mark Sisson seriously. He makes a lot of claims that can be backed up specifically, but he usually doesn't do the leg work to back them up, and he's made enough claims that I know to be verifiably false that I usually just ignore him[1]. He does suggest cutting grains of all kinds from our diet, but many of the problems he cites (insulin spikes) are remarkably different for different grains.

Google Books seems to have most of Good Calories, Bad Calories; searching within the book, there are 30 instances of the word rice; every one of those is either a quotation or Taubes specifically distinguishes between brown and white rice; in several places he implies that brown rice doesn't cause the problems that white rice does.

All this is adding together to reinforce my degree of confidence in considering brown rice an acceptable food.

[1] In the second article of his you linked to, he links to the pop-science coverage of an unpublished paper, cites their reported findings, and then outright not only dismisses but outright contradicts their conclusions with his own personal interpretation of what he wants their findings to mean.

Taubes specifically distinguishes between brown and white rice; in several places he implies that brown rice doesn't cause the problems that white rice does.

I believe brown rice has a lower glycemic index and glycemic load than white rice and it doesn't have some of the other negatives of grains like wheat (gluten for example). It's still high in carbs and low in nutrients relative to vegetables. It's probably one of the least bad grains but it's still something I prefer to avoid. Rice was pretty easy for me to give up anyway though, I never much liked it. The only thing I really miss is sushi / sushi rolls but I can still eat as much sashimi as I like so it's not too bad.

I also consider the "paleo diet" somewhat silly, not because I think there's a good reason for it not to be healthy, but because I think of it as impractical; how many of our modern fruits and vegetables, or even meats, closely resemble what we had available to eat more than 10kya? How many paleo practitioners will seriously evaluate making terrestrial arthropods a significant portion of their diet?

However, Evolutionary Health Promotion: A Consideration of Common Counterarguments so far seems to be the best argument I've found against brown rice, since it compares a Paleolithic diet against a Mediterranean diet, which contains quantities of unrefined grains, and finds moderate improvements for the former.

I also consider the "paleo diet" somewhat silly, not because I think there's a good reason for it not to be healthy, but because I think of it as impractical; how many of our modern fruits and vegetables, or even meats, closely resemble what we had available to eat more than 10kya? How many paleo practitioners will seriously evaluate making terrestrial arthropods a significant portion of their diet?

This seems to be rather an extreme case of making the perfect the enemy of the good. Sure there are differences between modern fruits, vegetables and meats and the pre-agricultural versions but they are a lot more similar to the pre-agricultural diet than grains or vegetable oils. Most paleo diet variants advocate trying to source foods that more closely match pre-agricultural staples - wild salmon, grass-fed beef, wild game, etc. - but I haven't seen much direct evidence of the benefits of that approach (although you can make inferences based on other evidence).

Anyway, I think diet may be one of those things where there is a big risk of other optimizing and I think it is quite likely that there is no one size fits all diet that is suitable for everyone (due to different genetic heritages and tolerances for things like gluten, dairy, etc.). It's an excellent place to apply some self-experimentation to see what works for you.

Paleo-type diets have worked well for me because they correspond closely to the kinds of foods I like to eat anyway and a major factor in the success of any diet is how 'deprived' you feel by the restrictions placed on what you can eat. YMMV but I think most people can probably benefit from eating fewer processed carbohydrates and sugar and more fruits and vegetables.

What's the name of the bias of judging messages based on the messenger? I suspect part of my problem is that everyone I've personally met who advocated the paleo diet was also into homeopathy and crystal healing, and didn't actually have any idea what paleolithic humans really ate.

The fact that we eat different things from our evolutionary ancestors should generate hypotheses about how dietary differences affect our health; two of the biggest changes to my mind would be a dramatic increase in starch consumption and an even larger increase in cooked versus raw food. However, I haven't seen evidence that either of those is specifically a negative change, whereas increase in sugar consumption has been conclusively shown to be a big dietary evil.

One of the reasons I'm so active on this sub-thread is that dietary changes over the past two years have made huge differences in my day-to-day life, but I feel like I could still do better. However, I'm not eager to abandon cooking my food, to start eating mammals, or to stop eating all grains, so I want to hear convincing evidence to do those type of things before I'm willing to self-experiment with them. Your point that I should perhaps not try those anyway (because of feeling too restricted) is well taken.

What's the name of the bias of judging messages based on the messenger?

I'm sure someone has discussed that here before .. the source heuristic? attribution/authority bias?

I suspect part of my problem is that everyone I've personally met who advocated the paleo diet was also into homeopathy and crystal healing, and didn't actually have any idea what paleolithic humans really ate.

Interesting. Most of the new age types I have met are into vegetarianism.

I discovered the paleolithic diet in the late 90's through this wonderful website devoted to it which I can no longer find. It presented a very convincing scientific case.

Humans evolved as opportunistic omnivores. Our bodies can run efficiently on a wide variety of diets. But nonetheless they are tuned to a certain range of conditions, and as you stray from that range random failures accumulate.

We changed suddenly in the shift to agriculture, and the archeological record shows that our bodies suffered - the move from meat rich paleolithic hunter-gatherer diets to neolithic agricultural diets is associated with a marked decrease in average height and overall ill-health. If I remember correctly, the average male european height was around 6'2"" 10k yrs ago, declined by half a foot in the neolithic age, and still has not fully recovered.

The evidence is also more specific. In every case where we have good specific evidence for biochemical problems in the diet, those problems are caused by dietary novelties. Examples: glycemic load, omega fatty acid profiles, potassium/sodium levels, nutrient load, etc - all of these are solved by the paleolithic diet.

So the real question is now that we have discovered all of these issues with the modern diet, how many other issues remain that we have yet to discover?

two of the biggest changes to my mind would be a dramatic increase in starch consumption and an even larger increase in cooked versus raw food. However, I haven't seen evidence that either of those is specifically a negative change, whereas increase in sugar consumption has been conclusively shown to be a big dietary evil

Starch turns into sugar near immediately - your saliva can almost do the job itself, but the transformation completes quickly in the stomach. You should look at a piece of bread and instead see a mound of sugar. Actually, the bread is more sugarey than table sugar - it converts more quickly into pure glucose from what I recall.

Starch is the staple of the neolithic agricultural revolution. Cheap energy for the masses. It works, and typical humans as of today already have adapted significantly, but not nearly as much as we should. These cheap fuels lessen our lifespan. Some specific humans have not had nearly as much time to adapt - some native american tribes in particular - and they react in a spectacularly poor fashion on modern diets.

You also mention cooking, but this is not a new neolithic change. There is some debate, but general consensus is that cooking has been around for much longer than 10k yrs, and we have had more time to adapt to it. Nonetheless, overcooked food has specific problems - mainly carcinogens.

jacob_cannell:

We changed suddenly in the shift to agriculture, and the archeological record shows that our bodies suffered - the move from meat rich paleolithic hunter-gatherer diets to neolithic agricultural diets is associated with a marked decrease in average height and overall ill-health. If I remember correctly, the average male european height was around 6'2"" 10k yrs ago, declined by half a foot in the neolithic age, and still has not fully recovered.

I've seen this claim many times from a bunch of people, but never accompanied by references to literature. Do you know what exact work it is based on, and how controversial it is in expert circles?

No. I was just recalling it from my memory, and my numbers are probably well off, but the effect seems to be widely acknowledged. It's difficult to really determine a trend like this - naturally height and bone stature can vary widely from region to region.

One summary here

some other data points from a paleo blog which show a less pronounced effect

what would be most illuminating would be to find an actual debate between paleontologists about this issue - that would probably have the highest density of info.

And on that note, other factors such as communicable diseases could also have played a role in decreasing average height and health of agriculturalists.

Be careful with things along the line of "the effect seems to be widely acknowledged", just because those effects that are are also often the ones least critically reviewed. I have also read the 6'2" height statistic, and never seen evidence that convinced me it was a reasonable average. I agree that a debate among paleontologists, especially one that completely omits mention of diet, disease or other explanatory causes, would be ideal to distill our best understanding of the facts. Then discussing diet, disease, etc. as possible explanations would move from facts to viable hypotheses.

Be careful with things along the line of "the effect seems to be widely acknowledged", just because those effects that are are also often the ones least critically reviewed.

Yes this can be true, and it is worth digging into. The 6'2"" figure is just a number from memory. From what little searching I did recently, I can't find much support for it.

The bone studies I have seen have a pretty wide dispersion of average heights. All show some trend for loss of average height around the dawn of agriculture (or even a little before), although it varies and is probably not as high as what I originally stated. This is probably so murky that it may not be worth drawing conclusions from.

A more interesting route could be to look at data from particular populations as they transition from hunter-gatherers to agriculturalists. There we could expect to see evidence for a significant loss of health initially followed by a massive selection effect, based on the health problems many humans from isolated tribes exhibit when exposed to the modern diet.

It just occurred to me that size is one of the easiest things to select for (which is why large dogs are an order of magnitude larger than small dogs), and agriculturists are in much less need of size than hunter gatherers. Maybe humans pre-agriculture were all naturally taller, but rarely reached their maximum height?

I believe that people living in developed nations usually grow to their maximum individual height, and that differences in height are almost entirely genetic among people with adequate nutrition. Since we've gotten taller as malnutrition has become less and less frequent, it seems reasonable to postulate that this is fulfilling potential rather than genetic drift. However, famines during the beginning of the agricultural revolution could have decreased the height of humans, since being larger didn't help one find alternate food sources (like it might have earlier in human history), but did require more sustenance.

There are probably a number of holes in this theory; despite humans breeding dogs to be vastly varying in size during this same period, I doubt that the time period in question is really quite enough to have altered human size naturally. I also think archeologists would be aware of this and be able to recognize changes even in malnourished size over the time periods being discussed.

Most of the new age types I have met are into vegetarianism.

You're not wrong...the paleo types that I've met are pretty crazy. I was hanging out with three of them a few months back (naked in a hot tub, 'natch), and all three considered themselves to be on the paleo diet, but also were raw-food only vegans, which I consider nuts, but corroborates your point about new-agers. Another that I met sometime earlier wasn't vegetarian, but was nearly as crazy as the other three.

all of these are solved by the paleolithic diet

I agree that they would be if the diet was carefully followed, but I wonder about how well people practicing the paleolithic diet actually approximate it. My biggest wonder is whether people work hard enough to eat only wild and grass-fed meats, which is necessary to balance out the omega fatty acid profile; otherwise meat is still bad for that. Nutrient load is also of interest to me, just because over the last several thousand years we've genetically engineered a lot of the nutrients out of our foods, though I suspect we haven't gone far enough for that to be a real concern.

I am also not convinced that our ancestors ate as much meat as is often claimed; present-day tribes certainly don't eat that much, even the most aggressive hunters, and many tribes subsist mostly on grubs and bugs for their protein. I'm not at all confident that too much meat is a problem in our diet, but I feel like the paleo diet is often a convenient excuse for people who want to eat a lot of meat, which is the same way I feel about Atkins. That doesn't speak to the actual worth of the diet, but it's worth keeping in mind as a potentially biasing factor.

bread is more sugarey than table sugar

Please stop repeating this popular myth. Glucose is a sugar, and starch is glucose, but that's very misleading. Glucose is naturally handled really well by the body, but fructose (and by extension, sucrose) is not. Sugar: The Bitter Truth is long but informative, although it has its detractors as well. However, both sources agree that glucose is just fine for you, while I don't think anybody (except lobbies, perhaps) seriously claims that table sugar isn't harmful.

I realize that cooked food has been around for a long time, but probably not long enough to induce significant evolutionary adaptations. What's really interesting to me is that animals---all primates, my dog, etc---prefer cooked food to uncooked, even if it's not a natural part of their diet.

I agree that they would be if the diet was carefully followed, but I wonder about how well people practicing the paleolithic diet actually approximate it.

Yes of course this is true, it can be difficult to follow. I believe in it in theory, but in practice is a whole other matter.

My biggest wonder is whether people work hard enough to eat only wild and grass-fed meats, which is necessary to balance out the omega fatty acid profile; otherwise meat is still bad for that.

Yes, although vegetable oils and fried foods can be just as bad. I take fish oil supplements to help at least.

bread is more sugarey than table sugar

Please stop repeating this popular myth. Glucose is a sugar, and starch is glucose, but that's very misleading.

I meant 'sugarey' in terms of the principle pure metabolic form of sugar, which is glucose. This is not a myth - starch converts so easily into glucose that you should think of it as glucose. Glucose is pure true sugar as far as your body is concerned. Thus starch is more pure 'sugarey' than table sugar. Ask a diabetic about the insulin response of starch.

You are discussing additional reasons why fructose is potentially even worse than starch/glucose.

Large glucose spikes are not "just fine for you" in the long term. There is significant mounting evidence that these large insulin spikes are a novel feature of our diet that we have not had adequate time to fully adapt for. In the long term it contributes to weight gain and the metabolic syndrome set of diseases.

Naturally we have varying degrees of genetic resistance, but a reduced insulin load is just plain healthier for everyone. This is no myth.

A key feature of the paleo-diet is low overall caloric density, low glucose, and low insulin load.

I realize that cooked food has been around for a long time, but probably not long enough to induce significant evolutionary adaptations.

I dont know about this. Some humans have significant evolutionary adaptations for grains and milk already. Cooking has been around for much longer, and we can expect significant adaptations for that. Indeed, if anything we have probably lost much of the robustness of our gut defenses as a result, but gained adaptations against potential ill-effects of cooking (carcinogens?)

Glucose does not evoke a metabolic response identical to fructose. Perhaps most importantly, glucose can be metabolized by all cells in the body, but fructose specifically stresses the liver. Quantitatively, fructose also causes much higher triglyceride and aldehyde production than an equivalent amount of glucose, and participates in glycogenesis much less efficiently; since glycogenesis is the strongest negative-feedback mechanism for dealing with excessive insulin, this ought to be a significant difference.

This is not to say that insulin spikes or high blood-glucose levels are healthful, but they can be regulated without eliminating carbohydrates from our diets. "Glucose is pure true sugar" as far as chemists are concerned, perhaps, but our body treats different sugars quite differently, and it is dangerous to conflate them.

I never disagreed with your point that fructose is worse than glucose. I haven't investigated it much, but I find it reasonable. I was merely pointing out that starch converts to mass glucose and mass glucose itself is not healthy. Not healthy at all.

Insulin spikes is one issue, but total insulin load itself is another. In theory you can control insulin spikes by eating a large number of smaller snacks, but in practice this is difficult for a variety of reasons - including not having any direct control over the rate of one's stomach emptying. As a diabetic could tell you (my brother is type 1), minimizing starch is crucial for reducing total insulin load.

We think of aging as measured by the passage of time, but the body does not measure time according to atomic clocks or rotations of the earth. Time in the body is measured in terms of biological processes.

Your heart's age is not measured in terms of seconds or minutes but in beats.

Your skin's age is measured largely in number of sun exposures and total sheddings.

In general the aging of most systems in the body can be measured in number of cellular reproductions and major metabolic cycles.

We know that general lifespan is measured more in total calories than absolute time. Total insulin response is perhaps an even more accurate measure, and at the very least it measures the differential aging of the metabolism of sugar regulation and energy storage in particular.

Reducing or eliminating starch is up there as one of the more important health improvements one can make. It helps reduce total calorie load, shift to a more ideal weight, and places less strain on your pancreas. High fiber carbohydrates are a different manner - but mostly what people mean by carbohydrates is typically starch.

We know that general lifespan is measured more in total calories than absolute time

So staying sedentary, while not overeating, should result in about a two-fold life-expectancy increase over an athlete that consumes about twice as many calories daily? Abstractly, that makes sense, but empirically we know that the truth is close to the opposite.

From what i remember about caloric restriction research, current empirical data does support the abstract theory, but I did not mean to imply some exact linear relation. From what I remember, exercise has health benefits but nothing near the scale of longevity increase you get from caloric restriction.

My impression is that caloric restriction has a threshold effect-- there's a mechanism that gets invoked when the calories are low enough, though it isn't a sharp boundary. Is this at all accurate?

This matches what I've read - that the body switches into different metabolic modes based on caloric abundance. Entire patterns of gene expression change.

Today we spend almost all of our time in the caloric-abundance high burn mode, and this accelerates metabolic aging.

However, I'm also thinking that caloric burn accelerates some aging processes by itself, but it is quite difficult to dissociate that from everything else - insulin, gene expression, etc etc

Would you accept a sub-linear correlation?

Person A = sedentary 60 kg female, eats 1500 kcal/day (which is not calorie restriction, but fairly reasonable for that body weight and activity level)

Person B = 60 kg female marathoner, eats 3000 kcal/day (high, but certainly possible)

Let's further assume that their diets are compositionally the same, so that B eats the same things as A in the same quantities, just twice as often.

Life expectancy difference? My money is on B to live longer. Yours?

some quick google searching on "athlete lifespan" looks to support my bet:

from this site concerning elite athletes more likely to fit that 2x caloric profile:

  1. What is the life expectancy of world class athletes?

Not so good. The average elite athlete will die by the age of 67. That is considerably lower then the 76 year life expectancy of the average American. Do you want to hear something that is really scary? According to the NFL Players Association, the average life expectancy of an NFL player is 58 years of age.

similar results for athletes in general

I can't really think of a study that would isolate for all the factors, but overall from what i've seen exercise is only healthy in moderate amounts, and taken to athletic levels the benefit is overshadowed by the higher caloric entropy.

Remember that heat is chaos, and we measure car lifetimes partly in mileage - so total entropy is a reasonable measure.

I suspect that a lot of the difference is accounted for by injuries, both major and minor. Football players, in particular, take a whole lot of abuse. A regular person exercising carefully at the gym won't have that problem.

Perhaps, but death from such injuries is extremely rare. I doubt it is any significant contributor to their death rate.

Burning off the calories through exercise is certainly more healthy than getting fat, but from what i've seen caloric intake is negatively correlated with lifespan in general.

With footballers, there's reason to think it's accumulated concussions rather than death from a single injury.

That's a good point and could explain why footballers have a lifespan 5 years or so lower than high end athletes and much lower than the general population.

It would be interesting to compare boxers/fighters based on weight class. They take alot of head beating and you can compare for different caloric intakes.

Sumo wrestlers would be another interesting data point - very high caloric intake and no concussions.

I suspect that one would find a significant caloric effect independent of concussions. Most footballers are quite big and have very high caloric intakes.

I'd expect QB's to live longer than typical linemen.

I cheated a little by mentioning that person B was a runner; I've seen those statistics, and they mysteriously don't apply to runners. Likewise, footballers (both gridiron and the real kind) have very high rates of arthritis, but runners do not. It seems to be a difference between what we were evolved to do and what we can do. Analogous to this discussion? Very likely :-)

One not-so-mysterious difference is muscle mass; long-distance runners don't develop much, whereas many other athletes do. Muscle is even harder to pump blood through than fat, so an extra 20kg of muscle is harder on the heart than 20kg of fat, which is sometimes cited as the reason that body builders so often have serious cardiac health problems (other reasons include large amounts of protein and cholesterol consumed, but I suspect you don't buy those explanations).

EDIT: I shouldn't stress my claim so much. I posted quickly because I'm busy with other things, but post-posting Googling seems to disagree with what I said; I haven't actually found anything I trust yet, but it does seem that moving from long-distance running to ultra-long-distance running may move athletes back into a risk category.

All else being equal and assuming A gets some exercise and is not fully sedentary, yes I'd bet on A.

Of course, the highest lifespan would be for caloric intake well below today's average.

As a diabetic could tell you (my brother is type 1), minimizing starch is crucial for reducing total insulin load.

Wait, what? As a diabetic I avoid starch (somewhat) to reduce blood glucose variability, but I'd never heard of "insulin load", and searching PubMed and Google for that phrase didn't turn up anything interesting. Starch only causes blood glucose variability if it's digested too quickly, as measured by glycemic index, which happens with things like breadcrumbs but not with things like bread. My understanding of the issue is that insulin matters only insofar as it interacts with blood glucose, glucagon and liver storage. If insulin had any direct bad effects, I would consider this surprising, and a big deal.

By insulin load i just mean total insulin. You diabetics know exactly how much you use in a given day, but regular people produce similar amounts automatically. The idea is that in the modern diet we consume too many calories mainly from energy dense foods such as carbohydrates which cause more insulin spikes and overwork the pancreas, eventually leading to destabilization and metabolic syndrome (type 2 diabetes and associated diseases). We haven't had enough time to optimize for these conditions.

There's also some recent evidence / claims that cooking (and meat eating) were in part responsible for the big jump in human brain size and intelligence.

That is interesting. I was aware that cooking opened up a whole new world of plant foods to us - most of our neolithic vegetable staples are indegistable or even poisonous until cooked. I was not aware that cooking significantly increases meat digestability. Although raw meats still seem much more digestable than raw neolithic staples. You can't eat raw grains or potatoes, for example.

Regardless of what other selection pressures may have been driving us towards larger brains, it's clear that hunting and cooking provided the resource advantage to support the quick increase. I imagine that the growth in complexity of language, culture and technology drove the selection effect.

It's interesting though that our hominid brains rapidly expanded up to about 100-200 billion neurons (the latter only in our larger brained neandrathal cousins) and then stopped. This is about the same upper neuron count we see in elephants and cetaceans.

all three considered themselves to be on the paleo diet, but also were raw-food only vegans

Raw-food veganism and paleo type diets are pretty hard to reconcile (though probably less so than regular vegan diets and paleo type diets). There are raw food animal diets but they are quite distinct from raw food veganism and are more similar to standard paleo type diets.

My biggest wonder is whether people work hard enough to eat only wild and grass-fed meats

This is probably the most difficult aspect of the diet to follow strictly due to the relative difficulty and expense of obtaining these types of meat and fish. That's why most of paleo type diets recommend Omega 3 supplementation and things like Omega 3 enriched eggs.

I feel like the paleo diet is often a convenient excuse for people who want to eat a lot of meat

This sounds to me like being stuck in the mindset that a diet has to be about deprivation or eating foods you don't particularly enjoy. There's no reason that healthy foods can't also be foods you like to eat. As a rule it seems that foods we are evolutionarily adapted to eat are both tasty and nutritious. It's the 'super-stimuli' of refined carbs, refined sugar and processed foods which tend to fit the stereotype of 'if it tastes good it must be bad for you'.

Glucose is naturally handled really well by the body, but fructose (and by extension, sucrose) is not.

Fructose and sucrose appear to be worse than glucose but it is false to say that glucose is naturally handled really well by the body, at least in the quantities produced by a typical western diet. Much of the harm associated with sugar and refined carbohydrate consumption appears to come from the over-stimulation of the insulin response and resulting blood sugar spikes and crashes caused by eating refined carbs and sugars. Bread is not 'more sugary' than table sugar but white bread does have a Glycemic Index that rivals table sugar.

Glucose is an important cellular fuel source and a key component of human metabolism but we are simply not well adapted to deal with the onslaught of glucose produced by consuming grains and refined sugars.

When I said that I considered a raw-food vegan paleo diet nuts, that's what I meant, that they just seem to difficult to reconcile.

This sounds to me like being stuck in the mindset that a diet has to be about deprivation or eating foods you don't particularly enjoy.

My statement wasn't an argument against meat; it was cautionary, and an appropriate response is to point out that I really like eating rice and beans, and should be equally cautious about arguing for their health benefits because of the stake I have in wanting those to be healthful.

My point about bread was that a 4oz piece of bread is not as unhealthy as a 4oz piece of hard candy. I don't consider refined carbohydrates, or even unrefined but purer carbohydrates (such as potatoes) healthful. I'm having a lot of trouble finding good sources to either back me up or discredit what I'm trying to say (the video linked to earlier is one of the former), but the impression that I'm under is that starch doesn't cause harmful insulin responses or large insulin spikes if eaten in the much less pure forms; that is, even significant quantities of carbohydrates are digested slowly enough to not cause problems if they are molecularly dispersed among protein, fats, or most especially fiber, and that this is a super-linear effect, more than would be explained simply by eating less of a pure carbohydrate.

but the impression that I'm under is that starch doesn't cause harmful insulin responses or large insulin spikes if eaten in the much less pure forms; that is, even significant quantities of carbohydrates are digested slowly enough to not cause problems if they are molecularly dispersed among protein, fats, or most especially fiber

I'm not too sure on what differences there are between starch and carbohydrates more generally but I believe it is true that insulin response is lower when carbohydrates are consumed in the form of foods with other components. This is the basic reasoning behind getting your carbs in the form of fruits and vegetables (which also provide plentiful fiber) rather than grains and tubers in paleo type diets. I've even seen it claimed that you are better off ordering fries rather than baked potato because the fat slows the rate of absorption but I don't know how well sourced that is.

Potatoes (not baked), bananas, wild rice and other similar medium GI foods are generally viewed as ok in small quantities by the paleo diet resources I've read. They still provoke significantly greater glucose response than things like green vegetables, fruits and berries however while having less overall nutritional value and not significantly greater fiber content.

People generally eat baked potatoes with butter, cheese, and/or sour cream-- not to mention meat.

For practical purposes, the carbs in a baked potato are very likely to be buffered with fat and protein.

I bet french fries could be better than a baked potato, but for that to be the case I think they would have to be unsalted (though the potassium in potatoes partially counteracts the salt) and fried in more healthful oils (probably saturated, since heat + unsaturated oil -> trans fats). This sounds like an experiment worth performing on myself; I like making french fries, and I could see how satiated I feel after a similar calorie-worth of french fries versus baked potato, and how I feel an hour later (more or less sluggish). I don't eat much potato in any form, but answering that question (at least with regard to myself) might be a worthy enterprise.

I personally think bananas are great, but it seems like they don't fit into the paleo scheme because they are so dramatically different than anything that existed in our ancestral environment. Wild potatoes are too toxic for humans to eat, even cooked IIRC.

I've also heard a plausible theory that a lot of the soil that food is grown on is mineral-depleted. Manure isn't going to solve the problem if it's from animals which have been fed food grown on mineral-depleted soil.

And I've wondered whether there are any important nutrients in insects.

What's the name of the bias of judging messages based on the messenger? I suspect part of my problem is that everyone I've personally met who advocated the paleo diet was also into homeopathy and crystal healing, and didn't actually have any idea what paleolithic humans really ate.

I'm not sure what the name of the bias is but I'd have probably had the same reaction as you if I'd seen the paleo diet associated with homeopathy or crystal healing. My impression is that it is over-represented in the contrarian cluster that is over-represented on LessWrong (atheist, libertarian, software engineer, not given to an interest in homeopathy or crystal healing) but that might just be some kind of selection bias on my part. I tend to associate raw-food / vegan eating with the nutty new-age fringe and have the same kind of negative associations with those diets as you describe for that reason.

I would imagine you could get pretty close to a paleo type diet without stopping cooking your food or eating mammals if you are ok with eating fish / seafood, eggs and cheese / dairy. It's pretty difficult to do it without cutting out grains however. I didn't find that too difficult (well, except beer) because I could always take or leave bread / rice / pasta anyway but I've heard from people who really enjoyed those foods that after a (sometimes difficult) period of adjustment they adapted to a grain free diet quite well.

[-][anonymous]40

Done that myself. It fixes a lot of problems: more energy, easier to stay full for longer on less food, the only "diet" I've ever tried that worked.

It's strange that something so "unnatural" seems to be good for the body along a variety of dimensions. (It's unnatural in the sense that animal products are expensive, and require a lot of biomass input, and so it's weird to make them a dietary staple whereas most of the world lives on grains.)

It's unnatural in the sense that animal products are expensive, and require a lot of biomass input, and so it's weird to make them a dietary staple whereas most of the world lives on grains.

It's natural in the sense that it seems to be closer to the pre-agricultural diet of hunter gatherers which represents a much larger period of time in our evolutionary history than the relatively recent rise of agriculture. That's the idea underlying the Paleolithic Diet which I've had a lot of success with.

In short, replace carbohydrates with animal fats.

Interesting. I've had little luck with that in the past, due to carb cravings. A few days ago, though, I googled "craving raisins and cheese" (my usual craving) and found a page that suggested this (and other sweets-cravings) were likely a need for tryptophan.

So, yesterday and today I've been taking some L-tryptophan and it seems to have knocked my carb cravings down a bunch; I've not had a compelling urge to eat anything sweet today as yet.

I used to desire carbs, but not particularly more than fats. My appetite for carbs has almost disappeared though. When I look at a plate of rice and beans, my body gives me the "that's not food. I don't want it" feeling.

Thanks for the L-tryptophan tip. I might get some and test it on some carb loving friends.

Mmmm... rice and beans...

Just to add to the anecdotal data, I've had the same experience upping animal fat and being able to be productive (mentally and physically) even without eating, and I work a physically demanding job at night, either of which alone can induce fatigue. I eat mostly butter, egg yolks, cream, coconut oil, and fatty cuts of meat like pork belly and fatty ground beef (epsom salts, mineral water and magnesium supplements take care of any muscle soreness).

[-]matt00

The thing holding me back from much more self-experimentation is a good data capture tool. Anyone have any recommendations (web, mac, iPhone)?

I especially agree with

I understand the mechanism explaining the effect very well.

When we're seriously in doubt, understanding mechanism can produce hypotheses to test. It's how we prune our search space.

In this case, we have two methods, (thanks for explaining what they were, by the way), both with plausible mechanisms to decrease mess. After even brief experimentation, it is obvious that method 1 has a specific flaw that prevents it from working well, and that method 2 works better because it suffers from no flaw as significant.

This last point is what is sometimes missed in (less careful application of) evo psych; plausible mechanisms are not necessarily well-understood mechanisms and probably can't be until they have been seen in action.

I'd like to know a) when we say "spilling seeds" what exactly do we mean? Spilling them into the pepper-pieces, or just spilling them over the table in general? With that in mind, what are the differences between the two methods? Your (old) way sounds possibly like my way, which I always thought was pretty good. But that might just be me being biased in favor of my own ingenuity.

I always start by cutting off the top of the pepper (high enough that it leaves a hole in the top-piece, because the little dip in the top goes lower than the place I cut at). This a) let's you pull out the core, probably in the "surgical" method you were referring to. But it does leave those little rubbery things that you need to deal with on the inside which sometimes have seeds left.

My old method was to cut a circle into the core end around the core and then pull out the core - typically leaving lots of seeds inside the pepper. Those seeds would get more widely distributed as I proceeded to cut up the rest of the pepper.

The method my wife taught me is to start cutting on the other end and make multiple parallel cross sections until I get to the core. The pepper now resembles a cup with a core in the middle. At that point I alter the direction of slicing and cut off the sides of the cup, leaving only the core. The advantages of this method are that a) I can clearly see the core when I am cutting around it, so I don't accidentally cut into it and b) I don't have to pull the core through a hole that tends to be too small for it.

I'm sure there are multiple non-stupid ways to slice a pepper. I'm quite bad at what we could call 3D reality manipulation and have to be explicitly taught obvious things. My father (also an engineer) is the same way.

I've had good luck with cutting off both the top and bottom 0.5cm of the pepper, which leaves me with a capless pepper cylinder containing the core. I can then slice out the core. Then before doing anything else with the pepper, I run water through the now-empty cylinder to rinse out any clinging seeds.

My favorite method is to cut it in half vertically/meridianally without cutting the core nor the stem portion, crack the two sides away from the core and discard it, then flatten and slice up the sides. With care, the core is undisturbed and intact, and you can slice up the rest quickly--maybe 15-20 sec per pepper to julienned, another 15 to diced.

However, if you are looking for pretty rings, which I think are often far too large to be bite size, my* method doesn't produce them.

(* 'my method' meaning as a friend/chef taught me, and is partially illustrated in Rombauer's "Joy of Cooking")

Following the theme of the post, perhaps the lesson is for one to look for sources outside oneself for criticism and alternatives so you can identify opportunities and metrics for improvement

My preferred method.