Friday, December 2, 2011

New Review Papers on Food Reward

As research on the role of reward/palatability in obesity continues to accelerate, interesting new papers are appearing weekly.  Here is a roundup of review papers I've encountered in the last three months.  These range from somewhat technical to very technical, but I think they should be mostly accessible to people with a background in the biological sciences. 

Food and Drug Reward: Overlapping Circuits in Human Obesity and Addiction
Written by Dr. Nora D. Volkow and colleagues.  This paper describes the similarities between the mechanisms of obesity and addiction, with a focus on human brain imaging studies.  Most researchers don't think obesity is an addiction per se, but the mechanisms (e.g., brain areas important for reward) do seem to overlap considerably.  This paper is well composed and got a lot of media attention.  Dr. Volkow is the director of the National Institute on Drug Abuse, a branch of the National Institutes of Health.  The NIH is the main source of biomedical research funding in the US, and also conducts its own research.

Here's a quote from the paper:

There is now evidence that comparable dopaminergic responses are linked with food reward and that these mechanisms are also likely to play a role in excessive food consumption and obesity. It is well known that certain foods, particularly those rich in sugars and fat, are potently rewarding (Lenoir et al. 2007). High-calorie foods can promote over-eating (eating that is uncoupled from energetic needs) and trigger learned associations between the stimulus and the reward (conditioning). In evolutionary terms, this property of palatable foods used to be advantageous in environments where food sources were scarce and/or unreliable, because it ensured that food was eaten when available, enabling energy to be stored in the body (as fat) for future use. Unfortunately, in societies like ours, where food is plentiful and constantly available, this adaptation has become a liability.
I think it's worth qualifying Dr. Volkow's statement that fat is highly rewarding.  It definitely is-- in the context of a low-fat or typical diet.  But in the context of a very high-fat diet, which is actually a reduced reward diet because it is restricted in carbohydrate, adding more fat does not increase the diet's reward value.  In that context, carbohydrate will be more rewarding, whereas adding more carbohydrate to a high-carbohydrate diet will not increase its reward value.  This is due to two factors: 1) carbohydrate and fat are both reward/palatability factors, and 2) food variety is a major reward/palatability factor (1).  This is probably the main reason why monotonous liquid diets are so low in reward/palatability value: they eliminate variety.  The main point I'm trying to make here is that I don't think fat is inherently fattening in humans-- it depends on the context.

Common Cellular and Molecular Mechanisms in Obesity and Drug Addiction
Written by Dr. Paul J. Kenny.  This paper covers a similar topic to the one above, but with more of a focus on animal studies.  Dr. Kenny is known for his work on D2 dopamine receptors in drug and food reward, and seems to be a growing authority in the field.  This is a very nice, fairly technical, review of the neurological and molecular mechanisms involved in both processes.  It was published in Nature Reviews Neuroscience, a high-profile science journal.

Is Fast Food Addictive?
Written by Drs. Andrea K. Garber and Robert H. Lustig.  I don't know Dr. Garber, but Dr. Lustig is known for his strong anti-fructose advocacy.  Dr. Lustig also believes that food reward/palatability contribute to obesity.  This paper makes the case that fast food may act on reward pathways to such a degree that it can be addictive in susceptible people, and contribute to obesity and poor health.

Metabolic and Hedonic Drives in the Neural Control of Appetite: Who is the Boss?
Written by Dr. Hans-Rudolph Berthoud.  Dr. Berthoud has been beating the reward-obesity drum for a long time now, and has written numerous review papers on the subject.  I haven't read this one, but I've read two of his previous reviews and enjoyed them.


Aeris said...

Thanks for the updates Stephan.

I'm surprised Lustig is interested
in food reward. I'll be sure to see what he's put out to see if I've written him off for no reason.

I was wondering if you have seen this article/study Stephan.

jeff said...

I wondered if you felt well-enough informed to provide an overview of the current state of research on inflammation. It seems to be highly connected to obesity and metabolic disorder, and you discussed inflammation byproducts in your discussion of the cytotoxic nature of excess energy, but I also feel it's used somewhat generically in the popular understanding, and my reading of Wikipedia left me wondering at the interaction of the inflammatory response with metabolic disorders.

THEGAP said...

The metabolic approach of obesity is wrong all mammals do have a brain and decisions about food are taken in the brain! All the metabolic processes are following these choices! And our brain is highly sensitive to all kinds of refined chemicals! As it is very good sense of daily observations of my behavior and those of the members of my family I would be very surprised that the agrobusiness should not be precisely aware of that...

dim sum massacre said...

The recent book 'The Compass of Pleasure' (by the neuroscientist David Linden)is a very interesting and informative overview of the current state of knowledge about the brain's reward system:

FredT said...

The conclusion to the cited paper

Our findings clearly demonstrate that intense sweetness can surpass cocaine reward, even in drug-sensitized and -addicted individuals. We speculate that the addictive potential of intense sweetness results from an inborn hypersensitivity to sweet tastants. In most mammals, including rats and humans, sweet receptors evolved in ancestral environments poor in sugars and are thus not adapted to high concentrations of sweet tastants. The supranormal stimulation of these receptors by sugar-rich diets, such as those now widely available in modern societies, would generate a supranormal reward signal in the brain, with the potential to override self-control mechanisms and thus to lead to addiction.

StanM said...

Roasted almonds I like to eat for pleasure a lot, potatoes with butter and some salt or beef live I like to eat when I am hungry but not for pleasure

Zorica Vuletic said...

This is why I advocate a moderate ketogenic diet. I'm not stating anything new here, just that I realize that such a diet can address the insulin problem as well as the dopamine problem.

Once you get an individual eating this way, it would be interesting to see if that person develops another non-food habit which is in an attempt to still raise dopamine receptor sensitivity. If this was the case, it would then strongly support the dopamine hypothesis of obesity and would then support proponents of the 'low food reward'.

Regardless, even if this hypothesis IS somehow correct in that obesity stems from the brain, I still don't believe in the prescription of low reward food such as the use of starches etc. Even if the primary reason really WAS reduced dopamine receptor sensitivity which lead to over-eating, the effects of elevated insulin are now the PRIMARY problem. The only way to 'fix' this is through moderate ketogenic diet. This diet will do two things: it will reduce insulin levels (as we all know) in the context of feeling satiated as well as restore dopamine sensitivity. The neuro-protective effects of ketone bodies in the brain has been well researched and supported though the use for epilepsy, Alzheimers and Parkinson's Disease.

Stephan Guyenet said...

Hi Aeris,

Thanks, I hadn't seen that and it is interesting.

Hi Jeff,

Yes, the connection between inflammatory signaling in the brain and obesity is one of the things I study professionally. I wrote about it in my "body fat setpoint" series.

kulimai said...

I'm still confused. I see the idea, I think it is excellent, but I do not see a theory. Surely one of the essential steps of a food reward theory should be a clarification of the concept of food reward. I have been following your very interesting blog quite a bit but I found only very cursory attempts at some sort of a definition.

You are explicit here that food reward is a contextual matter. So there is no such thing as the reward of food (component) X, food (component) X might have a reward factor that enters into the calculation of its reward (for individual Y at time t). So we don't know what the food reward factor is, what factors individuals (states of individuals really,-- due to adaptation,another major factor presumably, etc.) contribute and how we calculate reward from their interaction.

Obviously one does not expect a full answer but would you disagree that some systematic attempt at an approximation of the exact content of the reward concept would be necessary for there to be a theory and not just an good idea.

Have Fun To Be Young said...

Folks, it doesn't matter what workout you do it's really possible to LOSE 12 inches in 12 workouts with the "Fat Loss Zone" secret. Yep it works with cardio, Bootcamp, Weight lifting, Zumba, P90x, Insanity, Power 90, gym workouts, cardio classes, cycling, jogging, hiking, home workout videos ... anything folks, YES Anything!  After 10 years as a Personal Trainer and owner of Firm Body Boot Camp, the #1 Fat Loss Gym in the Country I learned the biggest mistake to fat loss is working out TOO HARD, and not in the your exact "Fat Loss Zone" zone causing your body lose muscle, slow down metabolism and store more body fat on your stomach, arms, chin and buns. I have mastered the art of Fat Loss and will teach you my #1 secret for FREE!

             If you're not getting results it's because 

        you are not using the 
"Fat Loss Zone" secret.

     Most people think working out extremely hard, sweat a ton and push yourself until you almost puke will deliver big results. They also think doing alot of cardio will shape, firm, tone the body and get them to lose weight. This is far from the truth, If you do any exercise TOO HARD your losing muscle ... NOT Body Fat.

Ryan Cowboy Ehmann

Cavegirl said...


You make an interesting point here and one that has rung a large bell with me.

I've been following a mainly ketogenic (ie 80 g daily or less of carbs) Primal diet for 2 years including regular 16-24 hour fasts.

I was a carb-loading vegetarian Ironman triathlete when I switched with a nice case of bipolar mainly expressed through mania.

Cut to 9 months into the diet switch and I could fast at will, had completely lost my sweet tooth and pretty much said goodbye to the bipolar.

I have never been a good sleeper but over the last 6 months or so my insomnia is proving more and more problematic and I now wake every hour and am often awake for several hours during the night. This without doubt messes with my appetite and creates carb-craving and general overeating regardless of activity level or hunger ... I go into graze mode and there isn't much I can do about.

So, I'm wondering if I'm developing a new behaviour to replace the now eradicated high carb/high octane regime I had before?

Certainly one to ponder.

RobR said...

I was thinking about reward and remembered the drug ibogaine has evidence of immediate cessation of opiate cravings. As it turns out there are tests being run on various components of the drug.

May lack some of the negative side effects in Rimonabant

As a funny observational study: Cannabis use linked to lower rates of obesity.

Kris said...

I completely agree with you, Stephan.

As a recovering drug addict, I can testify that junk food has many similarities with abusive drugs.

The cravings have the same kind of effect on thought processes, although not nearly as strong as with narcotics.

And I honestly believe that "junk food addiction", as I choose to call it, is one of the main reasons people today have such problems with cleaning up their diets, despite having a strong desire to do so.

7 Signs and symptoms you are addicted to junk food

Rap said...

Kulimai said:
"Obviously one does not expect a full answer but would you disagree that some systematic attempt at an approximation of the exact content of the reward concept would be necessary for there to be a theory and not just an good idea."

The ongoing problem. Even if some people grasp (or think they grasp) the concept clearly, it remains confusing for others. But there may be a way out. Many of the results being discussed on this site are, as I speculated in some previous comments, perhaps explainable as instances of habituation. From this perspective, it's not the taste or "palatability" of the food that is critical in, say, the cafeteria diet studies with rats. Rather, it's the utter lack of variation in the diet of the control group rats (just as it is the utter lack of variation in that potatoes-only diet or in that strange straw-sucking Nutrament study by Hashim).

Stephan has at times mentioned variability as a factor in food reward, and perhaps even made reference to the process of habituation (though I don't recall so), but perhaps variability is the primary factor. I did a quick search of the literature today, and the effect of habituation on preference for food items is in fact an established research area. For example, here's a recent study comparing one meal per day consumption of macaroni and cheese to weekly consumption for 5 weeks:
Macaroni and cheese hardly seems like a bland food item, at least subjectively (especially the cheese), but the daily meal pattern resulted in significantly reduced consumption, both for obese and nonobese participants (though they had predicted a possible difference).

And here's a recent review article on habituation and food:
Interesingly, at a glance, the review seems to make little reference to the food reward literature that Stephan cites. Are these more or less two separate lines of research, one of which emphasizes palatability (food reward theory) while the other emphasizes variability (habituation theory)?

A possible advantage of the habituation theory, apart from being based on a solidly established principle of learning, is that it makes little or no reference to "palatability" or the "rewarding" nature of food, and thereby largely avoids the subjective baggage that accompanies those labels. It also suggests that mere repetition of food items, both within and between meals, by itself may be an effective means for reducing consumption. This means that there may be no need to emphasize blandness in one's dietary efforts, which seems to be a problem for many.

Habituation theory may also account for certain phenomena that, from what I can gather, food reward theory cannot, such as the fact that people eat more when watching television or when suddenly presented a desert at the end of a satiating meal (the answer being that novel or distracting stimuli typically result in a weakening or removal of habituation, a process known as dishabituation). Finally, habituation is a normal process to which everyone is aubject, so there may be less need to construe normal overeating as a form of addiction, which may only be relevant in more extreme cases such as bulimia. I suspect that many people who gain weight slowly over the years do not pereive themselves as being driven by an addiction.

Keenan said...

Sounds like a cortisol problem, eat some carbs before you go to sleep. You crave carbs for a reason, your blood sugar is probably dipping to low which releases adrenalin, cortisol etc. and wakes you up. Seems like a pretty common problem among low carbers.

Cavegirl said...

Rip Clip

Mmm, I'd have to think about that one, I've never slept well, throughout my life, since a disturbing childhood in fact, it's a learned behaviour not to be too vulnerable.

I suffered with the sleep issues as a high carb vegetarian. After two years I'm totally keto-adapted, a fat burning beast as certain Primalists would say!

But it's an interesting thought, thanks.

STG said...

Please don't buy into speculation about your insomnia problem and what the solution is. Buy a glucose meter and test you blood glucose at night and morning (before and after dietary changes). Get real data and decide what to do from there. If you have blood sugar issues increased consumption of carbohydrates before bed could make things worse, or if you are not dealing with glucose problems increasing your carbohydrte intake make have results. If blood sugar regulation is not the problem then explore other possible variables. Other hormones can disrupt sleep (e.g., adrenal hormones, estrogen/progesterone. . . .). Stress can increase the likelihood of higher levels or cortisol and other stress hormones. Caffeine consumption late in the day or even moderate consumption in some individuals can impair sleep. You will have to be patient in working this problem.

Zorica Vuletic said...


I don't think fasting is suited for you. If you have glucose problems, even a low-carb diet is not enough if you go too long without some stability of glucose which is resulted from the long fasts.

Also consider increasing your fat intake and possibly evaluate your protein intake. You might not be able to tolerate too much protein.

Do not approach such woe as a 'primal' or whatever way of eating. You must look at a diet which is prescriptive. If you look at it as 'high fat, moderate protein and low carb' as opposed to 'primal/paleo' I think that will have more direction and will lead you to better health.

In conclusion, I do feel that the immediate step you can take is to eliminate the fasting. This is not always the best option for female body given our hormonal milieu. Perhaps if you did a lot of resistance training, you could get away with fasting, but overall I do not recommend this.

I would like to point you to Itsthewoo's blog. She really has a lot of understanding regarding this information. Just a few posts ago she explained how females tend not to be able to tolerate fasts like men can.

Of course, Peter of Hyperlipid always has useful things to say:

Another good blog to check out if you haven't already is Evolutionary Psychiatry of Emily Deans.

Let me know if you already read those blogs, and if you do I'll point to some other which you might not have known about before.

Travis Culp said...

I don't understand the sentiment that fat isn't inherently fattening. Long chain fats are transported via chylomicrons to adipocytes. Bizarre, supposedly ancestral, diets devoid of carbohydrates may not be fattening on a macro scale due to the emergency need to convert fat into glucose equivalents, but on the micro scale it's definitely fattening, as well as on the macro scale in a mixed diet.

The strangely common practice of eating a high fat diet coupled with regular periods of starvation and no carbohydrates may halt fat gain and/or cause bodyfat reduction, but the long chain fats consumed are still deposited, and thus the person is temporarily fattened by them.

I think biochemical facts are becoming distorted by the disordered eating of many supposed ancestral eaters. I sense that the original statement was intended to preempt rabid comments by the aforementioned fringe group but it's pretty clear that the most efficient way for a human to store energy in the long-term is to consume lipids.

Sue said...

Cavegirl, perhaps your diet went from one extreme to another and you need something in between. Dietary lifestyle must be contributing some part to the symptoms you are experiencing now.

Anonymous said...

These range from somewhat technical to very technical, but I think they should be mostly accessible to people with a background in the biological sciences.
Max Burn Funciona

Anonymous said...
This comment has been removed by the author.
Zorica Vuletic said...

@ Sue

The middle is bad in this case. I know many things are often 'better' in the 'middle' such as certain chemical levels...too much or too little with bad results.

HOWEVER: With diet, I think that this 'extreme' is better (ie: high fat, mod protein and low carb). Perhaps your fat intake is inadequate Cavegirl. Also, examine the types of fats too. Stick to nice saturated animal based fats. MCT oils are useful, however I don't think you need too much of those, they can be quite stimulating. Also, (as stated before) be careful of protein intake as that can be anxiety producing in some people.

Sue, if anything, a high fat diet with not too much protein and low carb is very calming and anxiolytic as evidenced by use in epilepsy.

I won't bother reference articles, since I might have already given links previously. I just know that one has to be pro-active and willing to self experiment.

Take this for example, one who moves from a war torn country to one which provides freedom. The country which provides freedom might ACTUALLY be very difficult to live in despite it not being ravaged by war. So, the individual might actually feel inclined to want to go back to their old home, because they at least know it and are familiar with it, and possibly still have family over there. So, the new country will 'hurt' despite it actually being better for the individual. So, within that new country, you have to learn how to tweak things so to speak so as to produce maximum benefit, not JUST the obvious 'safety' of life. This is just an analogy, and I just want to help. I don't want Cavegirl to be veered off on some 'safe starch' path or whatever JUST because there are a few things which simply need tweaking.

It's hard if one suffers mood related disorders so I just want to help. I don't know if she [Cavegirl] has any other attention and/or developmental related things to deal with such as ADHD or Aspergers, but anyway, I just want to see people getting the 'right' kind of help.

walia said...

This MCA project training is a golden chance for any individual seeking for a good start in the IT Industry. While undergoing the Project Training in Java, you will be able to explore your technical skills with innovative ideas and live projects. Our IT industry experts will guide and illustrate you on each of the modules and project requirements as well as they will help you master the other skills related to the subject.

MCA training in Chandigarh

Cavegirl said...


I have no wish to hijack this thread, I merely found Gladina's comment an interesting one but I appear to have opened a hornet's nest!

'Cavegirl' should link to my blog; I have done a lot of learning and self-experiment over 18 months, my regime is something around 55 - 60% fat (only animal fat and some olive oil), 10% carb and the balance protein, I have no alcohol, no caffeine and minimal dairy (just butter). I eat no grains/legumes, add no sugar.

I was a gestational diabetic; a higher carb diet always left me with big sugar dips, shakes, weakness and an inability to go long between eating.

No attention deficit! I read, on occasion, all the blogs mentioned here but now have so much contradictory information I tune into how my body is feeling and what makes sense intuitively, not what someone or another in the blogosphere says!

And I have no intention of adding more carbs, I've tried that and I swing back into problems - all to do with the dopamine/insulin/serotonin balances.

Inadequate sleep triggers all manner of hormone imbalances. I don't believe my insomnia is tightly diet-related, if you read my blog you'll find there are a myriad of other issues to work through.

Apologies Stephan, this is truly 'off topic' now!

Sue said...

you've always had insomnia and ketosis may be making it worse.

Cavegirl said...


And so might many other variables including the psychological spring cleaning I've been doing via Chinese traditional 5 element acupuncture ...

Ketosis is a natural state for humans, I can generate all the glucose my brain requires, internally, as do the traditionally eating Inuit.

Hélcio said...

As a bodybuilder for 27 years, I passed through many stages in food planning, dietting and eve drug abuse. Compulsive eatting is no different of gambling, sex, exercising, and drug.
I focused my life after 10 years of drug addiction in exercising, and after a detox threatment and rehab programs I am clean now.I had to make an Ibbogaine threatment to get free rom Cocaine and crack, and I am here to help anyone who needs information about.
visit me at
Keep Pumping!!

Jose Nanez said...
This comment has been removed by the author.
Jose Nanez said...

Hi, Stephan. Appreciate your work, as always.

I have a question though, a little off-topic. I was reading past exchanges in other blogs, when I found a comment of yours in which you stated that low carb diets have been found to be similarly effective to higher carb diets in terms of weight loss. The study you used as backing (, however, had a problem, because calories were limited in all diets, but not in the low carb one, and, apparently, carbohydrate levels where increased as time went by which affected the weight loss greatly ( This comes from a response that Peter (from hyperlipid) gave you. It just seemed odd that you would choose a study to support your claims that was be so inadequate. Sorry to bother you, but there's just one way to know, and that's asking.

Stephan Guyenet said...

Hi Jose,

Good point, I didn't realize the Mediterranean diet arm was calorie restricted as well. However, the study still does demonstrate that a calorie-restricted high-carbohydrate Mediterranean diet can equal the weight loss effect of a low-carbohydrate diet.

The low-carb arm was remarkably adherent, and maintained a similar reduced carbohydrate intake for the entire 2 year diet, therefore increasing carbohydrate intake cannot explain the fact that the low-carb group regained 1/3 of the weight they had lost initially.

Low-carb advocates like to point out that this diet tends to beat low-fat diets in weight loss trials. That is true. But it's like saying that an 80 year old man can outrun someone with two peg legs in a 100-yard dash. Neither diet is very effective in clinical trials.

If I was obese and the only choice I had was between low-carb and low-fat, I'd probably try the low-carb diet. But fortunately there are better options...

Stephan Guyenet said...

After 2 years of very good adherence on the low-carb diet in the Shai et al. study, participants were down ~5 kg (11 lbs) from a starting point of 91.8 kg (201.96 lb). Thus, they went from 91.8 kg (202 lb) to ~86.8 kg (191 lb), by cutting their carbohydrate intake in half for 2 years. This is not exactly blowing me away-- they went from very fat to slightly less fat.

The FDA requires that fat loss drugs produce a statistically significant decrease in body weight of 5% or greater to be approved. This works out to be 4.6 kg for a starting weight of 91.8 kg. So the low-carb diet arm just barely squeaked by the minimum
FDA requirement for weight loss drugs. Carbohydrate reduction can be very useful for a subset of people, but it certainly isn't a cure for the obesity epidemic.

Carnivore said...

The low carb group showed the best results despite the worst adherence rate at 24 months (78%) AND unrestricted calories. The Mediterranean diet resulted in slightly less weight loss despite being calorie restricted (1800 for men and 1500 for women). Doesn't that say something about the efficacy of a low carb diet? Keep in the mind that this diet wasn't really designed for massive weight loss:

"The low-carbohydrate, non–restricted-calorie diet aimed to provide 20 g of carbohydrates per day for the 2-month induction phase and immediately after religious holidays, with a gradual increase to a maximum of 120 g per day to maintain the weight loss".

So we're talking about a diet which was only looking to maintain the initial weight loss by gradually increasing carbs to 120g/day. If they really wanted to maximise weight loss, they could have given the subjects instructions to increase the carbs to say, only 50-60g per day. There's a huge metabolic difference between a diet that provides 120g carbs per day compared to one that provides 50-60g per day.

Low carb diets are generally the best and easiest way to lose weight for most people. Calorie restriction and low food reward diets may provide results too, but I wouldn't want to be an obese person following a diet where I was constantly hungry or one where I was supposed to eat bland food all the time, without all the seasoning and spices I want (we Asians NEED our food to be tasty). Long term, it's a recipe for failure for most, excepting those with extreme willpower. If the only concern is maximum weight loss, starving actually works best but it isn't sustainable either.

Alex said...

yeah 120g of carbs a day hardly even qualifies as low carb. I would call it medium carb. It's like calling a 2000 calorie a day diet low cal when the person's expenditure is 2100 a day. You really need to get down to under 80g to see good results.

Stephan Guyenet said...

Hi Carnivore,

In Shai et al., the LC arm ended up eating roughly half the typical amount of carbohydrate for 2 years. That qualifies as low-carb, although they definitely could have gone lower. If they had, they probably would have lost more weight as you said. However, the dropout rate and adverse effects would have increased.

The adherence rates at 24 months were "90.4% in the low-fat group, 85.3% in the Mediterranean diet group, and 78.0% in the low-carbohydrate group", and these differences were statistically significant. This suggests to me that people had an easier time maintaining the calorie-restricted LF and Med diet patterns than even this moderate amount of carbohydrate restriction. That does not speak well of the sustainability of LC for the average person.

Most obese people will lose fat on a very low-carbohydrate diet, but it isn't easy to maintain for most people, and the rate of adverse long-term outcomes seems to be pretty high (judging by the feedback I get from people). Personally, I can support it as a weight loss measure, but using it as a maintenance diet makes me uneasy. I can support moderate carb restriction as a maintenance diet if a person is clearly reacting well to it in the long-term (weight, energy, blood lipids, blood glucose). But not everyone reacts well.

Eating simple food, in my opinion, will be easier, more sustainable, have a lower risk of long-term complications and will probably be more effective for most people. There are so many different potential permutations: you could eat a simple breakfast and lunch, then have a normal dinner. You could eat small portions of typical food next to a large portion of plain starch such as rice or potato. Simple food does not mean eating bland food forever. You get to choose your own level of reward that suits your goals. You could keep spices, but reduce variety or added sugars/fats for example.

CarbSane said...

@Carnivore: You're confusing compliance with the diet with the rate of completion of the program. Stephan is right, the higher dropout rates for the LC group indicate that in "real life", this diet is more difficult to follow. But this adds non-sampling error to the mix here. The folks who completed LC were likely the most motivated of the bunch that were assigned to the diet, the others simply dropped out, whereas the LF group probably included some that were just motivated enough to hang in there but perhaps stretched the rules a bit more. That's one issue.

Taubes totally mangles the analysis of Shai. It's too long to repeat here, but I've blogged on this twice: and

It's important to realize that the LC group didn't up carbs until after 2 months but continued to lose up to 6 months ... after which they regained. Despite claiming slightly fewer carbs. All in all a good study to show what might be the best route for a majority but little else. And self-report free living studies are of little use other than gauging practical "real world experience" and efficacy.

Walker said...

A question that has been bugging me, I don't know if that has been brought up: what about bulimia and food reward? I mean the type that can ingest 4-5 thousands calories than attempt to vomit it. How would food reward work in this case?

Wouldn't flavor associations get out of whack if only a small % of calories end up being absorbed? The people who are mainly bulimic (as opposed to anorexic with bulimic episodes, say) are often of average weight, more so than with other type of EDs.

allison said...

This interesting report notes the link between sexual stimulation and the brain and its relation to food and drugs.

"At [female] orgasm itself, activity peaked in a reward area called the nucleus accumbens, an area that also responds, with less intensity, to cocaine, nicotine, caffeine and chocolate."

Carnivore said...

@Stephan: Thanks for your kind reply. I would say that if you want to judge the ease or difficulty of adherence to low carb, you'd be better off doing it from a collection of studies rather than just one. Here's one study where the low carb/Atkins group had the highest adherence rate:

There are a number of factors which determine what sort of a diet would be easy to follow for a particular person. Low carb is obviously going to be difficult for someone who prefers to eat vegetarian or someone who enjoys starch. For others, it really isn't that difficult or anymore difficult than a calorie restricted or low reward diet. The degree to which carbs are restricted obviously matters and for obese people, a longer period of strict restriction would help in losing more weight if that's the primary goal. However, it is possible to increase carbs to more moderate levels and still have good success, not just with weight loss but health markers in general. This aspect of low carb shouldn't be overlooked; it improves several aspects of health better than other diets. The people who face health problems when restricting carbs are usually the ones who are pretty active. They end up eating insufficient carbs for their activity level which results in problems like hypothyroidism, raised LDL, etc. Paul Jaminet has talked about glucose requirements being raised for active people and I think he has it spot on. Since obese people are usually sedentary, they're pretty safe even with long term carb restriction. Even Atkins himself recommends increasing carb intake after a certain time for weight maintenance.

If none of the people in these studies actually end up losing significant amounts of weight, it's because they're probably happy with modest results. There are only a few in the real world who end up losing over a 100lbs on ANY diet. Point is, low carb is the best choice for most and this is backed up by plenty of trials/studies. Whether we're actually impressed by the weight loss is another thing, but low carb consistently outperforms other diets despite being calorie unrestricted.

I take your point in reducing reward to meet one's goals. It's basically the same as my point about restricting carbs to a level that is sufficient for the required weight loss. I just think that if there were two options for me and one of them involved no compromise on the taste of my food, I'd take it without a second thought.

@CarbSane: I've already addressed the fact that adherence should be judged from more than one study. The one I linked to shows the highest adherence in the LC group.

People regaining happens in virtually every study. As time goes on, it is expected that the motivation will start to wear off and people will become less compliant and/or be looser with the amount of food they eat. This happens for all the diets, but the LC group still ends up with the best results, despite being calorie unrestricted. This is a very important point, because counting calories is a nuisance for most people and relying on hunger signals is much easier and more sustainable.

Finally, the whole point I'm making is that low carb is a very effective real world tool for weight loss and health improvement in free living subjects.

"self-report free living studies are of little use other than gauging practical "real world experience" and efficacy"

This is exactly what I'm gauging. I'm not theorizing about any magic abilities of the LC diet. Until there is sufficient data out there to show that low reward diets can bring about similar results and are as or more sustainable than LC diets, I shall remain skeptical. I like the theory itself, but I'm not so hot for the practical dietary advice it entails.

Sue said...

120g is low carb.

Rap said...

Here's an interesting twist that might work well for some. Twice a week low carb, low calorie with no restriction on other days.

Stephan Guyenet said...

Hi Walker,

Bulimia almost certainly involves food reward, because it is basically binge eating followed by vomiting. Binge eating is the classic example of "food addiction", i.e. out of control food reward. Binge eaters for the most part are literally addicted to highly rewarding/palatable food.

Your question is whether or not reward (particularly flavor-calorie associations) would become altered. I doubt it, because bulimics eat normally most of the time if I understand correctly.

Hi Carnivore,

I take your point about looking at dropout rates across multiple studies. However, in the study you cited the participants did not adhere very closely to the low-carb diet by the 12 month timepoint (46 -> 35% carb). Maybe they weren't dropping out because they were only half heartedly following the diet. Adherence in the other diet groups was poor as well.

The reason why I like Shai et al. is that 1) adherence was excellent, 2) the carb reduction was substantial, and 3) it lasted 2 years. It was really the best test of those diets that has been published so far IMO. Although it does leave some questions unresolved of course, e.g. what would have happened if everyone had eaten 50g carb per day for those 2 years.

I agree that when it comes to weight loss, low-carb generally beats low-fat (even when the low-fat arm includes advice to restrict calories), for the average person, according to clinical trials lasting up to 2 years. As far as how it measures up to other diets, I would say that's an unresolved question. People seemed to barely follow the Zone, Learn and Ornish diets in the A to Z study you cited.

You said "There are a number of factors which determine what sort of a diet would be easy to follow for a particular person." Yes, I agree. If someone finds that LC works well for them in the long term, I'm certainly not going to tell them to stop. I just think the effectiveness and sustainability of the diet for the average obese/overweight person are often overstated, and that leads people to stick with it even if it is clearly not benefiting them or even doing harm.

Alex said...


in regards to low carb diets, you said:

the rate of adverse long-term outcomes seems to be pretty high (judging by the feedback I get from people)

Could you go into a little more detail on this? What have you heard about from people?

Stephan Guyenet said...

Hi Alex,

The negative experiences I've heard about typically revolve around low energy/mood, deteriorating glucose control, greatly increased LDL cholesterol, and weight gain in some cases. I acknowledge that there are plenty of people who have seen improvements in most or all of those things with LC. The people who get into trouble tend to be those who restrict carb the most, according to what I've observed.

Travis Culp said...

Carbohydrate restriction lowers T3, leptin and sex hormones and increases cortisol. Incidentally, the same things are observed during outright starvation. It should not surprise people that there are frequent negative outcomes. Relative to being obese and sedentary, being less obese and sedentary is preferred, so for the intractably sedentary, carbohydrate restriction is a godsend. For those who have functioning appendages and are willing to use them, restricting fat intake to EFAs, eating a few hundred grams of carbohydrates and actually working out is far better.

Alex said...

Travis it's one thing to talk about your personal experiences and anecdotes but those are some very definitive statements you are making, do you have anything that backs that up?

Travis Culp said...

Why would the punishment for telling the truth and trying to help people avoid the frustration of never reaching the BF% they truly desire and the potential negative health outcomes of carbohydrate restriction be a bunch of work re-finding all of the studies I've read on the subject? You'll just have to trust me. I'm not personally invested in what you eat, so feel free to eat whatever you want. All butter or all Pop-Tarts, it's all the same to me. If you do a little digging, you'll see piles of evidence. Best of luck to you.

Sue said...

Alex you must be new to LC if you haven't read about low T3 and other symptoms of low carb diets.

Sue said...

Alex you said:
"You really need to get down to under 80g to see good results."

Do you have anything to back that up!!!!!

Travis Culp said...

What's better than loads of RCTs is simply taking your waking body temperature while you're restricting carbs and when you're eating a couple hundred grams per day. For me, the difference in body temperature is 1.5-2 degrees, which is a massive difference in perceived cold. If I recall, the reduction in T3 occurs due to a downregulation of the deiodinase enzymes that turn T4 into T3. Carbohydrate restriction, in essence, simulates a selenium deficiency.

CarbSane said...

@Stephan: Bulimics rarely eat "normally". That's the problem. It's feast or famine most of the time. I wonder sometimes, if you could isolate the bulimic in an environment free of mirrors, clad in sweats and provide them with a varied mildly hypercaloric diet for around a month or two, if this wouldn't cure many. People with eating disorders override their signaling for various reasons.

@carnivore: I get my diet comparison studies mixed up, but whichever one is the Dansinger study here was his takeaway message. Weight loss success is ultimately related to compliance, not diet composition. What is warped these days about low carbing is that folks promote equating a lack of deliberate caloric restriction with eating as much as you like (can!). This is why so many regain despite staying on the plan. In Shai, LC didn't come out on top after a year, and the Med group didn't regain significantly either.

Sue said...

Cavegirl, you'd rather believe some psychological spring cleaning is contributing to your insomnia rather than your diet which would contribute a hell of a lot more. Perhaps you have certain deficiencies like magnesium.
I believe ketosis can be therapeutic in a lot of cases but not all the time.

Deirdre said...

It sounds as though all of your diet modifications have not improved your sleeping issues throughout the years, which makes me wonder if this isn't so much of a nutrition issue as it is a brainwave issue. Have you considered neurofeedback training? Its been an effective therapy since the 1970s, biggest drawback being it's price (it could be very expensive). Anyway, thought I'd mention it to you. Best of luck - I hope you can find a solution!

Cavegirl said...

Sue and Deidre

The fact that a range of different diets over a 40 year period has made no difference to my disrupted sleep (which ranges from waking a few times a night to waking hourly and being awake for several hours - and has done throughout the last 4 decades) gives me the strongest intuitive feeling the fundamental issues lie elsewhere, at least in part.

I supplement with magnesium and vitamin D3, I live at 57 north!

My sleep issues as far as I can see were set up as a young child and are psychological ... I have been blogging about this for a while. There are lots of reasons why my psyche hasn't allowed me to sleep deeply.

Eating a low carb diet (generally under 100 g), which takes me in and out of ketosis has improved my mental health immeasurably, enough to give me the capacity to investigate further.

We did not evolve eating large amounts of carbohydrate, period. We do not require any carbohydrate, our body is equipped to convert what we need, our brains run more efficiently on lactate and ketones although they love glucose.

On standard diets I was ALWAYS hungry, often light-headed and had to eat, I become gestationally diabetic when pregnant, high insulin requirements knock out my dopamine systems and generate dysfunctional behaviour patterns.

I simply thought Gladina's original comment was interesting about the body generating a quest for a dopamine hit ... I don't get them through sugar highs any more, but disrupted sleep generates the kind of cravings that could lead to a sugar quest and the pleasure hit. I was just playing with a thought experiment ... perhaps my brain was cleverly tricking me into eating more sugar.

jyoti26 said...

good post
.Net training in chandigarh

Market Research Reports said...

it is nice blog shearing for health care
global biotechnology market

Deirdre said...

@ Cavegirl: While you acknowledged me in your response, I question whether or not you even read my post about neurofeedback training.

It's not food; It's wiring. Period.

If I were you, and have struggled like you have for years, I'd look into neurofeedback. You have nothing to lose at this point.

Carnivore said...

@Stephan: Yes, I am aware of the fact that people didn't strictly adhere to any of the diets, yet the dietary advice which said to restrict carbs produced the best weight loss and overall health outcomes. Dr. Gardner makes this point in his youtube presentation "Battle of the Diets".

"People seemed to barely follow the Zone, Learn and Ornish diets in the A to Z study you cited."

Another point in favour of the advice to restrict carbs! Dr. Gardner also makes the point that in the real world, the efficacy of a diet should be judged through the long term results it produces from a realistic level of adherence. This is obviously another point in favour of low carb.

The point is this: for quick weight loss, strict carb restriction has more benefits but it's harder to stick to this diet in the long term (neither is it recommended by Atkins). However, even moderate carb restriction produces great health results and decent weight loss and is easier to stick to.

The advice to eat plain foods is at best as simple as the advice to restrict carb-rich foods and no more easier to implement as a long term strategy. At worst, it implies that people have to eat bland food to see results (yes, not all the time).

"I just think the effectiveness and sustainability of the diet for the average obese/overweight person are often overstated, and that leads people to stick with it even if it is clearly not benefiting them or even doing harm."

To the contrary, I believe it's understated. The general public still tend to think of restricting carbs as something 'unhealthy'. It may be popular in a select group but not in the mainstream. The problem arises when some people stick with a version of low carb which is not right for them. Generally speaking, the more sedentary a person is, the more carb restriction can be tolerated. When active people who indulge in glycogen depleting activity restrict carbs severely, of course they will see poor results/health outcomes. Again, Paul Jaminet has talked about active people having increased glucose requirements and 'glucose deficiency symptoms' on VLC diets. For sedentary obese people, low carb diets are as safe as any other option, provided they implement it sensibly. In the rare case that someone's health actually declines despite following the diet properly, then of course that person shouldn't stick to it.


"Weight loss success is ultimately related to compliance, not diet composition."

Yes, I agree with the first part. As I said above, even relatively loose compliance to low carb produced better results than ANY other diet. Check out the presentation by Dr. Gardner "Battle of the Diets" which discusses the results of the study. If the composition of a diet influences compliance, then ultimately the success of the diet IS linked to composition.

"What is warped these days about low carbing is that folks promote equating a lack of deliberate caloric restriction with eating as much as you like (can!). This is why so many regain despite staying on the plan."

I don't think there's anything warped about the idea of eating to satiety and not having to count calories. Where are the 'so many people who regain' on the plan? Most people get great results in weight loss and health and maintain their weight loss even after increasing carb intake to more moderate levels. Even if a few pounds are regained, they are better off than when they started. (Jimmy Moore?).

"An In Shai, LC didn't come out on top after a year, and the Med group didn't regain significantly either."

In Shai, the LC group DID come out on top. It's clear from the weight loss charts. Med group did fine as well, probably because they too were restricting carbs, just not to the same extent.

Carnivore said...

@Travis: I think you are making judgements about carb restriction for athletes or active people based on your own negative experiences, which is fine.

However, it is quite obvious that athletes need more carbs, depending on their activity levels. When they don't get this amount, yes they suffer negative effects including lowered body temperature, less T3, etc. This doesn't imply in any way that the optimal diet for atheletes is high carb. It's still low carb, but not as low as for sedentary people. You just need to eat SUFFICIENT carbs for your activity level and all of the negative effects that you talk about will be avoided. The majority of your calories would still come from fat, with protein higher than for sedentary people. Here, Paul Jaminet discusses the optimal diet for athletes:

Carnivore said...

Another excellent post by Paul elaborating on 'Carbohydrates and the Thyroid':

Anonymous said...

Stephan, I'm interested in this comment of yours: "you could eat a simple breakfast and lunch, then have a normal dinner." I'm wondering how effective you would guess this to be, especially for someone without a lot of weight to lose?

I'm interested in whether lowering the reward value of my diet would help me lose about 10lb (I'm currently in the upper end of 'normal' weight range and would like to be a little leaner). I could implement a low reward plan for breakfast and lunch without too much problem, but I cook dinner for my family who have certain expectations. Our dinners are all home cooked from whole ingredients, so we're not talking industrial/supernormal levels of reward here, but neither are they really low reward.

Thanks for any thoughts!