Last week I addressed some criticisms of this series that I had seen at various places on the internet. This article represents some of my final thoughts on insulin and body weight regulation.
Are You a Cognitive Miser?
Consider this problem. Jack is looking at Anne, but Anne is looking at George. Jack is married, but George is not. Is a married person looking at an unmarried person?
- Yes
- No
- Cannot be determined
This problem is from an article in Scientific American about dysrationalia, a concept that describes where intelligent people think irrationally. The article also discusses a concept known as cognitive miserliness, which refers to people's tendencies to take short cuts in their thinking. These short cuts can then often lead to incorrect conclusions since the conclusions are based on incomplete information. In fact, cognitive miserliness can often lead to the phenomena of dysrationalia described in the article.
Have you determined your answer for the above problem yet? About 80% of people will choose the third option ("Cannot be determined"). However, that is not the correct answer. The correct answer is the first option ("Yes"). The puzzle doesn't say whether Anne is married or not; either she is, or she isn't. If Anne is married, she's looking at George, so the answer is "yes". If Anne is not married, Jack is still looking at her, so the answer is still "yes". Most people get this problem wrong because they don't take the time to go through all of the possibilities. Instead, they choose the easiest inference.
The Carbohydrate/Insulin Hypothesis: An Example of Cognitive Miserliness
The "carbohydrates drive insulin which drives fat storage" mantra is a perfect example of cognitive miserliness. It is a concept that is based on incomplete information regarding insulin, carbohydrate, and body weight regulation. Many people are drawn to this hypothesis because of its simplicity. However, its simplicity is exactly what makes it incorrect, because it does not address all the available data. It is a shortcut in thinking, based on a narrow view of insulin's effects on lipolysis and glucose.
One of the big problems with the carbohydrate/insulin hypothesis of obesity is that it does not address what happens over a 24 hour period. I briefly alluded to this in the first part of this series, and will expand upon it here.
Let's take alcohol as an example. Alcohol, when ingested, is known to suppress fat burning. This is because the metabolism of alcohol takes precedence over the metabolism of protein, carbohydrate, and fat. If one just looks at the suppression of fat burning, then one might predict that alcohol will cause weight and fat gain.
However, it is erroneous to make this prediction, because you need to address what happens over a 24 hour period, and what is happening in regards to overall energy balance. One study looked at the effects of alcohol on fat storage. The researchers compared two conditions. In one condition, alcohol was consumed in addition to meals, so that the subjects were consuming more calories than they were expending. In another condition, alcohol replaced some carbohydrate in the meals, so that the subjects remained in energy balance.
In both conditions, alcohol suppressed fat burning over the 6 hours that it was actively metabolized. When the subjects consumed alcohol in addition to their meals, fat balance over the day was positive, meaning the subjects gained fat. However, when the subjects consumed alcohol in place of some carbohydrate, there was no increase in fat balance over a 24-hour period. Although the alcohol suppressed fat burning over the 6 hours that it was metabolized, fat burning increased over the remainder of the day. This canceled out the suppressive effect of alcohol on fat burning. The authors stated,
Short term studies that fail to account for later readjustments of macronutrient balance can be misleading. We conclude that alcohol has a fat-sparing effect similar to that of carbohydrate and will only cause fat gain when consumed in excess of normal energy needs.
Just like with alcohol, one cannot simply look at the fact that carbohydrate ingestion will suppress fat burning during the time that carbohydrate is metabolized. Unfortunately, proponents of the "carbohydrates drive insulin which drives fat storage" mantra do exactly that. You have to consider the big picture and what is happening over a 24 hour period.
The other reasons why the carbohydrate/insulin hypothesis is an example of cognitive miserliness is due to all of the factors discussed in this series. There are numerous scientific observations that are inconsistent with the hypothesis. Also, the hypothesis, with its narrow focus on insulin, fails to address the myriad of other hormones and factors that are all interacting simultaneously to regulate body weight.
The Finale
The bottom line is that insulin does not deserve the bad reputation that it has been given, and the "carbohydrates drive insulin which drives fat storage" mantra is wrong. To summarize:
- Insulin suppresses appetite; it does not increase it
- A high carbohydrate diet does not cause chronically high insulin levels
- Protein is insulinemic, and in certain cases, can be just as insulinemic as carbohydrate
- Contrary to popular belief, glucagon does not "cancel out" the suppression of lipolysis by insulin when protein is ingested
- The insulinemic effects of protein are due to a direct stimulatory effect on the pancreas, and not because the protein is converted to glucose
- The combination of protein and carbohydrate can produce greater insulin secretion than either one alone, yet high protein, moderate-to-high carbohydrate diets are very effective for weight loss
- Very high carbohydrate diets have been demonstrated to produce weight loss when people are in an energy deficit
- Dairy is extremely insulinemic, just as insulinemic as white bread, yet does not promote weight gain in the absence of an energy surplus. This is supported by a very large number of studies, including animal studies, observational studies, and randomized controlled trials.
- Insulin is not required for fat storage
- Insulin levels are not predictive of weight gain or weight loss in the majority of prospective studies
- Exenatide restores rapid phase insulin release in diabetics, yet causes weight loss
- The effects of insulin injection cannot be compared to normal physiological insulin release, since amylin is co-secreted with insulin from the pancreas
- Insulin mainly functions as an inhibitory hormone rather than a storage hormone, acting as a brake on many important physiologic processes
- A type I diabetic without insulin becomes hyperglycemic because of overproduction of glucose by the liver, not because insulin can't get into cells
- Insulin enhances the uptake of glucose into cells, but is not required for it
- Insulin regulates blood sugar after a meal both by stopping the liver from producing glucose and enhancing glucose uptake into cells.
- In a fasted state, insulin regulates blood sugar by controlling glucose production of the liver, not by affecting the uptake of glucose into cells
- You cannot simply look at the temporary effects of insulin on lipolysis and glucose storage. You have to address what is happening over a 24-hour period; body fat will not increase if there is no overall energy surplus.
Don't be a cognitive miser. Insulin is not the bad guy.
Did Martin finally shut up? He must be that stalker/troll that harasses all the nutrition blogs. Another dumb lonely old man that should run on out and eat a bullet for being such a ****ing loser.
He should have simply stated a question and stated his thoughts in his very first post rather than playing word games. Worthless POS. Hey Martin Levac, do us a favor and stop wasting internet badwidth you **** for brains.
I blocked Martin from being able to post on the site a long time ago. He’s only one of 3 people I’ve blocked. Fred Hahn and Razwell are the other two.
James any ideas on how I’m still losing weight by raising my energy consumption to 5000 calories a day? I’ve increased my energy intake, but I’m still losing weight. I’d like to know how this fits in with your models.
I might want to add that it was not muscle I gained. I looked puffy, inflammed, and was tired all the time and was constantly sore even taking protein powder (which I was accounting for in my calories).
I did better when I tried the milk only diet (where I only consumed whole milk and multi-vitamins). Which got me to my lowest weight ever of 195, but I was constantly hungry (I’m guessing something was missing from the diet).
James, I think you insulin is a straw man you’ve setup. While you statements seem mostly consistent (except where pointed out otherwise by Todd and Martin), you don’t actually come out and state a thesis. Saying insulin is not bad for you is a worthless statement. There is pretty compelling evidence as admitted by yourself that insulin plays a role in regulating lipolysis and fat oxidation. Thus for someone who wants maximal weight loss would it not behoove themselves to maximize their insulin sensitivity via diet? Signaling mechanism in the body, particularly one such as insulin, affects multiple things at… Read more »
Hi James, Excellent series on insulin! If insulin’s major role in lowering hyperglycemia is turning down hepatic glucose production where does that leave peripheral insulin resistance? For example in hyperinsulinemic clamp studies they will typically tout increased glucose uptake in the muscles as a measure of improved skeletal muscle insulin sensitivity. Is this a faulty or insignificant metric or are skeletal muscle insulin sensitivity and peripheral insulin resistance not related? My understanding after reading Sonksen’s piece was that the unchecked lipolysis was the real problem with peripheral insulin resistance? Also, Sonksen mentions that ketosis can prevent glucose from entering the… Read more »
Straw man. Did you or did you not say: “The removal of insulin causes hyperphagia.” What this means is that hyperphagia is an effect. Did you or did you not say: “Nowhere have I claimed that hyperphagia causes obesity under all conditions.” What this means is that you did say that hyperphagia causes obesity under certain conditions. What this further means is that hyperphagia is a cause. Did you or did you not say: “In fact, higher insulin levels were associated with greater fullness.” This means satiety and thus cessation of eating is an effect. Did you or did you… Read more »
Addendum. With your argument that when insulin is associated with satiety, you mean to say that insulin is an anorexiant. Anorexia is the opposite of hyperphagia (polyphagia in wiki). Effects, not causes.
Martin, your logic in this discussion is highly distorted and flawed. For example, Did you or did you not say: “In fact, higher insulin levels were associated with greater fullness.” This means satiety and thus cessation of eating is an effect. No, I said that higher insulin levels were associated with greater fullness in that particular study. Correlation does not equal causation. Please read this post on the cum hoc, ergo propter hoc fallacy. Did you or did you not say: “The direct demonstration of cause/effect is not a “faulty inference.” Etc… This means you know precisely what the concept… Read more »
Did you or did you not write: “direct evidence that insulin suppresses appetite.” If that’s not equivalent to “causes satiety”, shoot me and call me crazy. Yet did you or did you not say: “No, I said that higher insulin levels were associated with greater fullness in that particular study.” That looks like denial. James, you said it. It’s not my fault if it doesn’t fit, it’s your fault: You said it. You said: “I never claimed that variation of insulin within a physiological range will affect satiety.” I never said you did. Here’s what I tried to show you… Read more »
Martin, there is no point in continuing this discussion with you, and here’s why. First, you continue to take quotes out of context and create contradictions where there are none. This is your style of argumentation: James: When you press on the accelerator pedal and the car engine is on and the gear is in Drive, the engine will rev and the car will move forward. James: When you press on the accelerator pedal and the car engine is on and the gear is in Park, the engine will rev and the car will not move forward. James: When you… Read more »
James, you keep calling “straw man” and “that’s irrelevant” as if doing so will win you arguments. It doesn’t. It merely changes the goal posts continuously so that your arguments are now made valid. Like you’re both a player and the referee and you’ve thrown out every other player so you’re the only player left that can score a point. You’ve thrown me out James. You win. Congratulations James, for what that’s worth. You are now the sole member of a very exclusive club. Others who agree with you are just standing on the sidelines not trying very much to… Read more »
James, you keep calling “straw man” and “that’s irrelevant” as if doing so will win you arguments. The purpose of a debate is not to “win.” If your intent is to win this, that is more indicative of a dogmatic position rather than an evidence-based position. Calling something a strawman is extremely relevant to the conversation. If you are going to construct faux arguments that I never made, in the interest of giving yourself something that you can easily dismantle (never mind that it is not the position I held), then I will call you out on it. Being able… Read more »
Linearity is not required for a dose response to occur. Furthermore, the variation in insulin resistance across various tissues changes insulin’s perceived effects accordingly. The linearity could very well be affected by this mechanism. For example, we could perceive an increase in plasma insulin merely because certain cells have suddenly become resistant in response to a non-insulinogenic stimuli. Butter would be such a stimuli. The point is that we have here a perceived effect of insulin without the direct action of insulin itself or of an insulinogenic agent. Further, ignoring insulin when talking about calories is like ignoring physiological mechanisms… Read more »
Linearity is not required for a dose response to occur. Of course it doesn’t. But the problem is that in your simplistic argument comparing one extreme (no insulin) to the other extreme (supraphysiological doses of insulin), a linear or near-linear relationship is assumed. Dose-response relationships are quite complex. For example, with some substances, there are U shaped relationships where you see certain effects up to a certain dose, and then attenuated effects at higher doses. In other cases (like the health effects of alcohol), there are benefits to moderate doses and deteriments to high doses. There are also logarithmic relationships,… Read more »
Riddle me this.
y1 = x1 + x2 + x3
y2 = 3×1 +2×2+ x3
y3 = x1 + x2 – x3
z = 6×1 + 4×2 – x3
Does x1 have an effect on z if it can be independent controlled? Does it cease to have an effect on z if it cannot be independently controlled?
I’m clearly oversimplifying here, but just because a biophysicist can model body composition based off feeding habits does not mean that insulin does not play a role in fat storage.
What kind of argument is that?
“What type of argument is that?” It’s not an argument that I ever made. I never claimed that insulin does not play a role in fat storage. There are numerous hormones that play a role in fat storage. But that does not mean that you need to account for the effects of insulin (or any other hormone) when modeling body fat change. You do not need to know the inner workings of a black box to be able to accurately model its inputs and outputs. Genes also play a role in fat storage. So does that mean I would need… Read more »
I need to add one more comment to this, because again you are constructing a strawman. The purpose of my entire series is to show that the carbohydrate/insulin hypothesis simply fails to explain a large number of scientific observations, from the effects of dairy on body weight to the fact that insulin levels do not predict changes in body weight. This is not “ignoring” insulin…it’s showing that trying to structure a diet based on its effects on insulin is misguided because the underlying hypothesis is simply wrong.
James, you said that removal of insulin in type 1 diabetics results in hyperphagia. Your argument about calories stops making sense here. That’s because in addition to the hyperphagia, there’s emaciation. And the addition of insulin causes fat accumulation. How can hyperphagia cause emaciation? And how can the reversal of condition, i.e. the addition of insulin, cause fat accumulation? You use evidence that refutes your own arguments. And you expect us to pay for that stuff in the future?
I guess everybody’s got pay his rent.
James, you said that removal of insulin in type 1 diabetics results in hyperphagia. Your argument about calories stops making sense here. That’s because in addition to the hyperphagia, there’s emaciation. Actually, it makes perfect sense. The removal of insulin removes insulin’s inhibitory functions on appetite, which results in hyperphagia (the same effect is created in rodents when you block central insulin receptors in the brain). However, the removal of insulin also removes insulin’s inhibitory functions on proteolysis and lipolysis. You get “runaway” proteolysis and lipolysis, which results in the emaciation. No amount of food intake is able to make… Read more »
Indeed. You said “the removal of insulin also removes insulin’s inhibitory functions on proteolysis and lipolysis. You get “runaway” proteolysis and lipolysis”. Yet you claim that “Insulin is not required for fat storage”. By which mechanism, prey tell, can fat accumulation, let alone excess fat accumulation, is made possible without the action of insulin? Do I need to remind you that you said “No amount of food intake is able to make up for the runaway proteolysis and lipolysis.” You said “Hyperphagia doesn’t cause the emaciation.” Of course it does not. What you claim is that hyperphagia causes obesity. But… Read more »
You said “Hyperphagia doesn’t cause the emaciation.” Of course it does not. What you claim is that hyperphagia causes obesity. But there’s a problem with this. If hyperphagia causes obesity, and if removing insulin causes hyperphagia, then removing insulin should cause obesity. Martin, once again you have created a straw man. Nowhere have I claimed that hyperphagia causes obesity under all conditions. Hyperphagia is defined as an abnormal appetite and increased consumption of food. If this increased consumption/appetite is met by an increased energy expenditure (or a dramatically increased excretion of glucose/ketones in the urine as in the uncontrolled type… Read more »
Let me remind you of what you said: “The direct demonstration of cause/effect is not a “faulty inference.” If one injects insulin centrally and it suppresses appetite, versus no suppression when only the vehicle is injected, then this is direct evidence that insulin suppresses appetite. When you add the evidence that knocking out insulin receptors in the brain results in hyperphagia, as well as the removal of insulin in type I diabetics results in hyperphagia, then you really cannot get better evidence that insulin is an appetite suppressant.” If the rule you established is good enough to determine cause and… Read more »
If the rule you established is good enough to determine cause and effect with hyperphagia, it’s good enough to determine cause and effect with a caloric deficit. So what’s the cause of our caloric deficit? That’s right, the removal of insulin. Just like with our hyperphagia. Martin, this is third strawman that you have constructed. Yes, the removal of insulin causes a caloric deficit, because it results in massive excretion of glucose and ketones in the urine, along with elevated energy expenditure due to hyperglucagonemia and massive protein oxidation. I thoroughly explained this in my last comment; why you ignore… Read more »
The complete lack of insulin, or chronic hypoinsulinemia as it’s technically called, is merely one end of the spectrum of insulin. The other end is chronic hyperinsulinemia. Everything in between is merely a variation on this theme. Indeed, correcting hypoinsulinemia only a little will result in only a little correction of the emaciation. Accordingly, correcting hypoinsulinemia too much will result in an exaggerated correction of emaciation, i.e. they’ll grow fatter than normal. Indeed, insulin is used in bodybuilding for its anabolic qualities, on top of MGF, IGF, growth hormone, testosterone, etc. In other words, it’s just one end of the… Read more »
And would you believe it, more insulin means more growth, and less insulin means less growth. This is where you are incorrect. The effects of insulin are not linear across its concentrations. This is perfectly illustrated where the absence of insulin results in an increase in energy expenditure in vivo, yet reducing insulin by reducing carbohydrate does not increase energy expenditure in vivo. And would you believe it further, bodybuilders know that injecting too much insulin makes you fat. And why would they know that unless it was true? What, are they injecting a different substance mistakenly believing it’s insulin?… Read more »
Hey James, you’re wrong. Alcohol acts exactly like fructose in the liver. I’ll give you a few minutes to figure this one out.
Hey James, you’re wrong. Alcohol acts exactly like fructose in the liver. And, like fructose, the adverse effects (increases in triglycerides, fatty liver accumulation, etc) only occur at high doses beyond a certain threshold. The threshold for fructose is estimated to be anywhere from 60 grams to 100 grams per day in humans. And even these effects are not completely independent of energy expenditure. For example, a 50% fructose diet (which is a very high fructose intake) does not impair exercise-induced body fat loss in rodents (who have a much higher capacity for de novo lipogenesis from fructose than humans).… Read more »
I’m sorry I wasn’t clear enough. Please try to refute your own argument instead of trying to confirm it all the time. Maybe then you’ll understand why I said that alcohol acts exactly like fructose in the liver.
Martin,
Claiming that alcohol behaves like fructose in the liver (in which they are similar) is not refutation of my argument. Given your tendencies to produce straw men, non-sequiturs, and other fallacies as shown in your other comments, I have my doubts that your thinking regarding alcohol and fructose is sound as well.
Just think it through will you instead of refusing to play. Why is it relevant that fructose acts exactly like alcohol in the liver? I’ll give you a hint, it has to do with fat burning suppression.
Just think it through will you instead of refusing to play. Why is it relevant that fructose acts exactly like alcohol in the liver? I’ll give you a hint, it has to do with fat burning suppression. Martin, you are providing an example of the “cognitive miserliness” that I discussed in this article. Yes, fructose, like alcohol, takes precedence in oxidative pathways. But, as I stated in the article, this fails to address what is happening over a 24-hour period. It also does not consider the energy status of the body or liver glycogen status. As I stated in the… Read more »
In the mean time, here’s another question regarding that study. Did they isolate insulin? What I mean is did they compare two diets of equal calories but of different insulin load? No, they merely compared two diets of differing calories and differing insulin load. How then can we conclude that it was all about calories and not all about insulin? In order to conclude anything, we have to isolate both insulin and calories. So, we have to compare two diets of equal calories but of differing insulin load, and two diets of equal insulin load but of differing calories. Otherwise… Read more »
What study are you referring to? I referenced a study on alcohol and total fat balance to illustrate that alcohol, despite the fact that it inhibits fat oxidation, does not result in positive fat balance if participants are in energy balance. Whether insulin was measured is irrelevant. This study was to show that, when considering the effects of a particular macronutrient on fat oxidation, you have to consider what is happening over a 24 hour period and not just during the time the macronutrient is being oxidized. I then compared this to how people view insulin. People blame insulin for… Read more »