Tuesday, July 5, 2011

Stoned On Fat?

Neuroscience researchers at UC Irvine, , led by Daniel Piomelli, have just announced results of a study done on rats, in which they found that ingestion of fatty foods stimulates the release of endocannibinoids, chemicals that activate the same receptors affected by THC.  According to the UC Irvine news release:

"The process starts on the tongue, where fats in food generate a signal that travels first to the brain and then through a nerve bundle called the vagus to the intestines. There, the signal stimulates the production of endocannabinoids, which initiates a surge in cell signaling that prompts the wanton intake of fatty foods, Piomelli said, probably by initiating the release of digestive chemicals linked to hunger and satiety that compel us to eat more."
In other words, rats eating fats had the munchies.  Feeding rats carbohydrate or protein did not have this effect.  Piomelli offers an evolutionary explanation:

"Piomelli said that from an evolutionary standpoint, there’s a compelling need for animals to consume fats, which are scarce in nature but crucial for proper cell functioning. In contemporary human society, however, fats are readily available, and the innate drive to eat fatty foods leads to obesity, diabetes and cancer."
The study results will appear this week in the online edition of Proceedings of the National Academy of Sciences

This finding dovetails with the reward theory of overeating, suggesting that animals including humans get more immediate endogenous drug-like reward from eating fats than carbohydrates or proteins, partially because fats supply more than twice as much energy per gram as carbohydrate or protein.

Piomelli apparently also serves as Director of the UCI School of Medicine’s Center for Drug Discovery & Development and hopes to find a pharmaceutical solution:

"The findings suggest it might be possible to curb this tendency by obstructing endocannabinoid activity – for example, by using drugs that “clog” cannabinoid receptors. Since these drugs wouldn’t need to enter the brain, they shouldn’t cause the central side effects — anxiety and depression — seen when endocannabinoid signaling is blocked in the brain, Piomelli noted."
So there you go, a pharmaceutical solution to the obesity problem.   Just throw a monkey wrench into the intricate and poorly understood symphony of neurochemicals.  No need to worry about side-effects, right?

How about teaching people to recreate the ancestral environment instead?  What did he say?  "Fats are scarce in nature."  By "nature" he means in the ancestral environment

Could this be why humans need only about 20-25 g of essential fats daily, compared to ~50-60 g of protein and at least 150 g of glucose?  Does it make sense that human macronutrient requirements would mirror the relative availability of nutrients in the ancestral environment and diet?

And are things really that different in the agricultural food supply?  I mean, although fat seems abundant in industrialized nations, does this reflect nature, or human intervention?  After all, agriculture is part of nature.  Does agriculture produce more fats, proteins, or carbohydrates?  If you look at the world at large, at the entire human food supply on the planet, is fat relatively abundant, or relatively scarce, although concentrated in certain locations?  How about protein?  Carbohydrate?

Why would evolution favor a system that offers an animal a higher immediate reward for eating fats than for eating protein or carbohydrate? 

Would this reward system be more advantageous in an environment where fats were easy to obtain, or in an environment where fats were hard to obtain?

In other words, would nature make it more highly rewarding to eat something available frequently, with little effort, or something available only infrequently and with great effort?

From another angle, which would this system help most:  an animal that had a continuous supply of fats, or one that had an intermittent supply of fats?


nothing91 said...

Thanks for the rat study, Don. Let's go right ahead and extrapolate to humans without hesitation. :-)

"As I read this claim that foods rich in fat rarely occur in nature, I wondered if people who think this way have ever seen a bison, with its big hump of back fat? A beaver? A duck? How about a salmon? Mackerel? A seal? A whale? A dugong (sea cow)?"

- Don Matesz, 2009

"If they think fat doesn’t just grow on trees—maybe they forgot about palm trees, coconut trees, pecan trees, walnut trees, etc.?"

- Don Matesz, 2009

"Funny how such statements about the scarcity of fats in nature get repeated frequently, but I don’t recall ever seeing any popular pundit pen the following parallel statement: 'While the modern human diet supplies lots of refined sugar and grains, such foods actually don’t occur in nature.' Did I miss it?"

- Don Matesz, 2009

"Fat provides more than twice the energy of carbohydrate, making it THE high energy fuel of choice. Any animal that could learn how to get a hold of fat would have a huge selective advantage over those that did not."

- Don Matesz, 2009

gpc said...

maybe it's simply reward for doing the right thing...

Don said...


Where did I extrapolate? I quoted Piomelli, and then asked some questions.

Did you extrapolate?

Do you have any evidence that this mechanism is not present in humans?

Did you know that many studies on humans show that humans also get reward value from fats?

And what exactly is your point in quoting my errors from the past?

Please show me some evidence that what I said in those quotes is more accurate than what I said in this post.

I'm not afraid to say that in all those quotes, I made a very significant mistake guided by a very common cognitive error, namely confirmation bias. I pointed to a few examples of foods that may contain fat, and implied that those few examples proved that fat is abundant in nature. In not one of those quotes did I even once provide evidence that nature actually produces more fats than carbohydrates.

If fat is abundant in nature, why would nature have to give even a rat a huge reward to get it interested in going after it?

If trying to train kids or animals, do you give them rewards for doing things they can easily accomplish? Or only for doing things that they rarely or never accomplish, unless rewarded?

Don said...


Please also show me evidence that humans REQUIRE more dietary fat than carbohydrate.

One frequent mistake I made repeatedly in the past was to ignore established facts of human nutritional biochemistry, thinking that studying a few junk piles from the past (archaeology), other species (e.g. H. habilis and H. erectus), imaginative accounts of prehistoric people (various stories of human evolution), and the lifestyles of certain outlier groups of modern humans (hunter-gatherers) would tell me more about how the modern human body works. I think I was wrong to do that.

I used to think that if biochemists, medical researchers, and nutritionists would pay more attention to anthropology and archaeology, they would get a more accurate picture of human nutritional requirements.

Now it seems to me the other way around: if archaeologists and anthropologists had more accurate knowledge of modern human nutritional requirements, they would get a more accurate picture of the ancestral evolutionary environment and lifestyle.

I previously wanted to use the very hard to study past to try to understand the more easily studied present. Now I would rather use the more certain knowledge of modern human nutritional physiology to come to some understanding of what is much less certain or easily studied (prehistoric diets).

Anonymous said...

Awesome series of posts Don. Keep up the good work and we will keep ignoring the trolls.

Wout Mertens said...

Where does the number of 20-25g essential fats come from? With essential fats, do you mean the Omega-3/6 PUFAs?

I thought the optimal amount for those is around 4% of calories or 10-15g? (not that I have a study for those numbers either, it's just what I read the most)

Wout Mertens said...

BTW, this mechanism may well be present in humans - eating fatty food sure makes me feel good!

However, the article doesn't mention anything about the negative feedback loop that is certainly present in humans - i.e. satiation. I don't see why you would need to stop the reward process provided you eat only foods that don't mess with the reward measurement system.

nothing91 said...


"Do you have any evidence that this mechanism is not present in humans?"

Are you asking me to prove a negative?

"Did you know that many studies on humans show that humans also get reward value from fats?"

I don't think anyone would deny that fats provide reward value. But what's fascinating is that a year ago you would have interpreted this as meaning fats are healthy -- which is perfectly logical. Nowadays, however, since you're on a "fat is bad" kick, you interpret it as meaning fats are rare in nature. The latter interpretation is clearly more of a stretch -- but it fits into your current world-view, so it must be true.

"And what exactly is your point in quoting my errors from the past?"

Primarily to show how sure you (think you) are every time your position changes.

"I'm not afraid to say that in all those quotes, I made a very significant mistake guided by a very common cognitive error, namely confirmation bias."

You're doing the same thing now. You just don't realize it yet. You only focus on evidence that you can interpret in a way that supports your viewpoint. Your viewpoint has done a 180 but the confirmation bias hasn't changed at all.

"If fat is abundant in nature, why would nature have to give even a rat a huge reward to get it interested in going after it? If trying to train kids or animals, do you give them rewards for doing things they can easily accomplish?"

Again, this "reward means non-abundant" stuff is bias-induced stretching if I've ever seen it.

What if animal fat is abundant but harder to get than plant food (hunting vs. gathering)? Logic would suggest animal fat would produce a reward in this case in order to encourage us to go after it rather than taking the lazy route and simply gathering.

This scenario could, of course, mean that more fat would be consumed than carbs. The extra reward of fat would certainly make this plausible. (As long as you don't make baseless leaps from reward -> abundance, that is.)

Your issue, Don, is a lack of an open mind. You get your sights set on a given viewpoint, and this determines how you "see" the evidence. You're now able top see the error of your ways in the past, but you're blind to the fact that you're doing the exact same thing now.

Malena said...


I would rather say that after following one dogma (or sect) with negative results, you actually can be quite certain that this time you actually got it right. Based on your experience. Your own body.

I also bought into the idea that fat makes you thin and health, however, only with negative results. But I have been extremely lean on high carb low fat without even trying. I gradually swopped diet, unfortunately. Right now I'm asking myself why. Depression, lethargy and some fat pounds on my body was the result. Instantly with a low fat diet I see changes.

I think to be open minded is to try. Full heartedly. For enough time to understand what is happening. Not a mere week, months or a year. After several years you can actually reach a long-term conclusion. Then, persons who really have an open mind are willing to change when circumstances and experiences point at another direction. It is those who always stick to the same dogma that are blind. And as Don obviously is not blind, as proven by his recent postings, I am sure he will be open to new thoughts and new experiences and continue to change opinions in the future.

If there is one virtue in life it is to be able to recognize your faults and mistakes and change your opinion. Over and over again.

JasonB said...

I'm curious about how often starchy carbs occur in nature vs. fats. We're in an unstable and unnatural system wherein we grow significanly more carbohydrate foods (corn, rice, wheat, oats, soy, etc.) than would be found naturally occurring.
I won't extrapolate that to mean that more ruminants would be around because I think we're forcibly raising more than would naturally occur, too.

Robert said...


Realize that two different people may respond differently to the same diet i.e. diet is not a one-size-fits-all issue. Brilliant minds like Alan Aragon and Lyle McDonald have been saying this too-simple-for-people-to-accept-truism for years. Just because you found what works for you means that it is a universal truth.

Moreover, I think that your diets have more in common than you recognize (e.g. minimally processed, exclusion of certain "neolithic agents," etc.). Moderation has worked best for me and I don't think rigidity is the best approach.

Also,--and I could be completely wrong here--wouldn't a high reward be a signal that it is good for us in naturally-occurring amounts (cocaine is not "natural")? Wouldn't that suggest that our ancestors who "enjoyed" fats fared better than those who didn't? Of course, there may be a Red Queen-ish competition between food and brain, so this could be useless speculation.

Random question: Don, if breast milk is perfect for babies, does that have any implications for adults' diet?

Thanks for your thoughts.


Don said...


I think breast milk can tell us some things about optimal human nutrition, but only if you keep in mind that it is meant for a human that doubles its body weight in 180 days.

Low carb advocates focus on its high fat content, but when you compare human milk to other species' milk, human milk is not particularly high in fat, and has almost half the fat of cat milk. The fat is to support growth of tissues requiring fats.

Even though it supplies the infant with a high protein diet (~1.7 g/kg), human milk has ~7 times more carbs than protein.

If a baby doubling its weight every 180 days only needs 1.7 g PRO/kg, how can any adult need that much protein. Not even bodybuilders can double their body mass in 180 days, so 1.7 g/kg is way above i.e. double adult requirements.

Human milk has the highest carbohydrate content of any species listed here:


As I said, it supplies ~7 times more carbohydrate than protein. If you look at milks of carnivorous species (cats, dogs), they have low carbohydrate, high protein, high fat compared to human milk.

Also, in general, the faster the animal grows, the lower the carb content and higher the protein and/or fat. This suggests that the carbs in milk are not to promote growth, they are to fuel the body especially the brain.

Thus, a comparison of babies to adults and human milk to that of other mammals suggests that human adults need way less than 1.7g PRO/kg, way less fat (unless you want to double your body weight every 180 days), and something like 6-7 times as much carbohydrate as protein.

Perhaps I will devote a whole post to this topic.

Sue said...

"The findings suggest it might be possible to curb this tendency by obstructing endocannabinoid activity – for example, by using drugs that “clog” cannabinoid receptors. Since these drugs wouldn’t need to enter the brain, they shouldn’t cause the central side effects — anxiety and depression — seen when endocannabinoid signaling is blocked in the brain, Piomelli noted."

Doesn't this suggest that if you block endocannabinoid (EC) can cause depression and anxiety. So with fat increasing EC that's a good thing particular in people with depression and anxiety.

B said...

Well I did have a great day yesterday, maybe it's because I ate 250g of fat (and 109g protein, and 140g carbs)?

I only weigh 105 lbs though. Less underweight than before a high-fat high-nutrient diet, but still underweight.

Anand Srivastava said...

I disagree that the requirement of fat is 25-30gms. The carbs will provide the deficit. Similarly no carbs are actually required if there is enough protein to cover the deficit.

I agree with Stephan that a higher carb diet is PROBABLY better than a low carb diet iff you do not have any glucose metabolism issues. I have read from Peter that rats generally are not able to handle fats well, and get fat when given unrestricted access to them. So the study given sounds logical but would not apply to us.

Wout Mertens said...


Actually both mink and donkeys have higher lactose content than human milk in that table (though not by much).

Furthermore, since lactose synthesis seems to require glucose as the first step, wouldn't it be possible that people eating a diet rich in carbs would convert more to lactose? The table was published in 1995.

It would be very interesting to find out what the variability of human milk composition is across the world.

Malena said...


If you mean 250g butter, 109g chicken and 140 banana you are not eating much for being underweight and wanting to change that. If you actually mean pure fat/protein/carbs -grams, then your digestion probably isn't working. Food allergies? UC? Chron's? "fast metabolism"? (which actually in many cases mean the food passes the your intestines too fast).

For all those problems I would DEFINITELY cut out all grains (as you probably already have) but also all milk products. I've been in contact with several people who had constant diarrhoea and/or ulcerative colitis (for like 20 years), to be totally cured when cutting out the milk (and gluten-grains and soy).

There could of course be other reasons for a bad nutrient/energy uptake, but in 99% of the cases they are food related.

Robert said...


Thanks for your response. I've the "Perfect Health Diet," which uses the macronutrient composition of breast milk as evidence for supporting their macro ratios.

Also, I'm not sure it's fair to say that a high fat and protein diet makes you fat in light of all the anecdotal and scientific evidence to the contrary. Again, I think it will vary among individuals.

sandra said...

Hi Don,

I appreciate your blog and this discussion...

others have asked if the high reward caused by ingestion of fat could be simply because of our need for it. Could you comment on that taking the following articles into account? Both seem to indicate that carotene is not well absorbed/converted to A (esp. when dietary fat is low...)


"Worryingly, younger women are at particular risk (of vitamin A deficiency)," Dr Lietz added. "The older generations tend to eat more eggs, milk and liver which are naturally rich in vitamin A whereas the health-conscious youngsters on low-fat diets are relying heavily on the beta-carotene form of the nutrient." Many health conscious elders also strive to follow a low-fat diet and thus avoid vitamin A rich foods


"We confirmed the highly variable extent to which ß-carotene provides vitamin A, even in identical diets. Some could achieve adequate vitamin A nutritional status from ß-carotene alone, but 45% would not. We found no biochemical markers (eg, total fat, protein, or cholesterol) in our small group of relatively uniform healthy women that were associated with low response..."

"The low absorption values that we measured mimic what might be expected of ß-carotene considering recommended American dietary practices (ie, increased consumption of fruit and vegetables and reduced dietary fat)"

Don said...


As Piomelli said, the reward system is to reinforce a search for adequate essential fats. That doesn't mean that it is designed to encourage a high-fat diet. Again, the reward encourages the animal to do something difficult to do in a natural setting.


"In the average collecting area of an Aka Pygmy group in the African rain forest, the permanent wild tuber biomass is >4545 kg (>5 tons)"


Wild grasses also produce enormous amounts of starch in seeds.

"Over large areas of Africa people once obtained their basic subsistence from wild grasses....Gathering grains from grasslands is among the most sustainable organized food production systems in the world."




The table shows mink and ass milk with 0.1 % greater lactose content...not enough to say they are higher because measurement error could account for the difference.

20 g of fat includes 17 g for omega 6 and 2 g for omega-3, as specified in the current NRC RDAs. Some research suggests lower is possible.

Don said...


A person can get or stay lean on high protein and high fat IF they can stay in neutral or negative fat and caloric balance. Some will some won't.

Don said...


I searched pubmed for any article documenting vitamin A deficiency due to a low fat diet with only adequate intake of carotenes as a retinol source. I found none.

A 2009 consensus conference on the topic concluded that B-carotene is an important vitamin A source for humans.

You only need about 4% of calories as fat to have adequate absorption of carotenes according to Bullock and Rosenthal's Pathophysiology text book.

I used to think that that AJCN study showed that 45% of people don't convert carotenes to vitamin A. However, when I read it more carefully, I noticed that 7 of 11 subjects had been taking retinol supplements before onset of the study, an average of 9300 IU daily, i.e. intake adequate to maintain high stores in individuals with low requirements. Also, the authors state:

"Starting on day 10 and throughout the remainder of the study, each subject received a vitamin A supplement of 1250 IU (375 RE, or 1.31 µmol as retinyl palmitate in cod liver oil) every other day at breakfast."

Giving people with previously adequate or even excessive intakes of vitamin A a regular dose of vitamin A virtually ensures that some will NOT respond to B-carotene by converting it to vitamin A since the body does the conversion on demand. For some people, 1250 IU preformed vitamin A every other day might be a sufficient intake to suppress conversion of B-carotene to retinol.

To test conversion of B-carotenes to A you should use people with history of very low to no intake of preformed vitamin A, and run the study with 0 intake of preformed vitamin A for the duration (to maximize the demand for carotene conversion), neither of which applies to this study. In other words, this study was almost by design destined to produce the outcome reported. It was poorly designed.

Again, I have been unable to find any medical report of even one case of a person eating a low fat diet, with adequate B-carotene from whole foods, but no preformed vitamin A, resulting in vitamin A deficiency.

The LM review you cited mentioned several factors that could be responsible for low conversion, including, possibly, obesity.

In any case, even if a person requires retinol, it is completely unnecessary to eat a high fat diet to get adequate intake. Just 2 tsp of cod liver oil will provide 50% of the RDA.

But beware, excess vitamin A may promote osteoporosis if vitamin D intake is low.


Which may suggest a reason that some people have low conversion of carotenes to vitamin A with relatively low vitamin A intakes as in the AJCN study. Perhaps when vitamin D is low (common), the body in its primal wisdom reduces conversion of carotenes to A, to prevent development of a harmful imbalance of A and D.

Don said...

Here's an example of a study on Filipino kids with low fat diets and whole foods rich in carotenes. With mimimal fat, only 7 g (6% of energy) per day, a diet rich in carotenes doubled vitamin A stores over a 9 week period.


This study fulfills my criteria: no previous intake of vitamin A supplements or preformed vitamin A from foods, and only 12 mcg dietary retinol daily (less than 3% of requirements) during the 9 weeks of the study. So, under conditions of very low fat intake and very low retinol intake, humans successfully convert carotenes to retinol enough to enlarge stores, even during growth and development periods of high demand for vitamin A.

Anonymous said...


"Low carb advocates focus on its high fat content, but when you compare human milk to other species' milk, human milk is not particularly high in fat, and has almost half the fat of cat milk. The fat is to support growth of tissues requiring fats."

This is flawed logic!!! Do kittens drink their mother's milk for up to 2 years??? Human babies breastfeed for much much longer than any animals. Basically an animal who is dependent on their mother's milk for a relatively much shorter time get their fat dosage in a "condensed" time period. Human babies however need a steady supply of fats over an extended period and hence the relatively lower fat content.

Don said...

Sorry, in last comment I made a mistake. During the Filipino study I cited, on 5 days per week the preformed vitamin A intake was 82 mcg per day 5 days per week, and 141 mcg on the other 2 days. So on most days the intake of preformed retinol was 14% of requirements, and twice weekly it was 23% of requirements. In any case, these children had a demand for vitamin A that was met by carotenes with a very low fat intake.

Don said...


How is it flawed? Humans grow slower so they need less fat to support that growth, so the milk is lower in fat. Kittens grow faster so they need more fat in their milk to support the growth rate. The faster the animal grows, the more fat it needs. So humans need a smaller dose of fat per day, but over a longer period of time. I don't see how anything you said shows flawed logic in what I wrote; in fact it simply confirms what I said.

Fat has 9 kcal per gram, which makes a high-fat diet (breast milk) ideal for promoting growth in animals with high metabolic rates (infants have a metabolic rate 2-3 times that of adults), so Nature puts a lot of fat in milks, and more the faster the animal grows, but a high-fat diet is not necessarily ideal for adults which by nature have lower metabolic rates, especially relatively sedentary modern adults.

nothing91 said...

Good post by Emily Deans here: http://evolutionarypsychiatry.blogspot.com/2011/07/endocannabinoids-fat-and-rats.html

As I suspected, the fat used in this study was corn oil rather than something that actually exists in nature.

You forgot to mention that, Don. :-)

Kiran said...

This study says that people with certain common SNPs on the BCMO1 gene have slower metabolism of β-carotene into retinol.


Wout Mertens said...

I found this interesting review paper from 1986 discussing human milk composition. It turns out that the total distribution of macronutrients is pretty stable.

Fat content is at 4-5%, more than cow milk (3.25% according to nutritiondata), but the composition varied greatly with diet.

The lactose content is stable except in two poorly controlled studies - one had a high fat, low carb diet that made lactose levels _rise_.

Likewise total protein levels are stable.


So babies get 39% of calories from carbs, still leaving 55% from fat and 6% from protein.

Hobson said...
This comment has been removed by the author.
Gadfly said...

Equivocate much, Don?

Fat is fat is fat, right? Right?!?

And the assertion that a reward system would indicate that fat was scarce is laughable. It could just as easily indicate that the substance is essential to healthy functioning. As always, the ratio of kaka thrown to kaka stuck is all about volume.

Anonymous said...

Great blogging tips. Have more trouble managing my blog than I do with writing blog posts. This helps out immensely. back pain