Saturday, October 1, 2011

The Case for the Food Reward Hypothesis of Obesity, Part I

Introduction

When you want to investigate something using the scientific method, first you create a model that you hope describes a natural phenomenon-- this is called a hypothesis.  Then you go about testing that model against reality, under controlled conditions, to see if it has any predictive power.  There is rarely a single experiment, or single study, that can demonstrate that a hypothesis is correct.  Most important hypotheses require many mutually buttressing lines of evidence from multiple research groups before they're widely accepted.  Although it's not necessary, understanding the mechanism by which an effect occurs, and having that mechanism be consistent with the hypothesis, adds substantially to the case.

With that in mind, this post will go into greater detail on the evidence supporting food reward and palatability as major factors in the regulation of food intake and body fatness.  There is a large amount of supportive evidence at this point, which is rapidly expanding due to the efforts of many brilliant researchers, however for the sake of clarity and brevity, so far I've only given a "tip of the iceberg" view of it.  But there are two types of people who want more detail: (1) the skeptics, and (2) scientifically inclined people who want mechanism.  This post is for them.  It will get technical at times, as there is no other way to convey the material effectively.


First, some definitions.  One of the problems with food reward is it's defined differently by different people, even researchers, and often it isn't defined at all.  As defined in previous posts, I use the term food reward to refer specifically to the motivational value of food, i.e. its ability to reinforce behavior.  For example, acquiring a taste that causes a person to seek out the food in question more often.  This is how some, but not all, researchers define the term.  Others use the term "food reward" to refer to both the motivational and the palatability value of food.  Palatability refers specifically to the enjoyment derived from a food, also called its hedonic value.  Palatability and reward typically travel together, but not always.

The food reward hypothesis of obesity states that the reward (reinforcing, motivational) and hedonic (pleasure, palatability) value of food influence food intake and body fatness, contributing to the development of obesity.  Quite a bit is known about the central nervous system (CNS, i.e. brain) circuitry that underlies reward and hedonic processing, and how this circuitry influences food intake and body fatness.  Most of the research on reward and hedonic processing began with the study of drug addiction in animals and humans, however these circuits evolved to guide behaviors that enhance fitness in the natural environment such as those relating to food and sex, and research also focuses on these processes.

Although food reward and food palatability typically occur together, they are not the same thing.  In the 1990s, Dr. Kent Berridge proposed that behavioral and CNS responses to food can be divided into "wanting", corresponding to the reward/motivational aspects of food, and "liking", corresponding to the hedonic/palatability aspects of food, and this framework has been a good fit for the evidence since then (1).  These two elements can be separated from one another experimentally, and sometimes in daily life as well.  One example is a person who is addicted to a drug despite no longer deriving pleasure from it.  This is a case of strong "wanting" without "liking".

Brain Circuits and Neurotransmitters Underlying Food Reward and Hedonic Processing


The fact that the reward and palatability value of food can be experimentally dissociated implies that they are mediated by distinct CNS circuits.  Indeed, although reward and hedonic circuits overlap to some degree and influence one another, each system has its own unique circuitry and suite of chemical neurotransmitters.  CNS regions important for reward processing include the orbitofrontal cortex, amygdala, nucleus accumbens, dorsal striatum, ventral tegmental area, substantia nigra and the lateral hypothalamus (2).  Dopamine signaling in these areas is a particularly important component of reward, but has little impact on hedonic processes (3).  Dopamine isn't so much a pleasure chemical as it is a reward/motivation chemical. 

CNS regions important for processing food-related hedonic information include the brainstem, pons, nucleus accumbens, ventral pallidum, amygdala, insula and prefrontal cortex (4).  Note that these overlap somewhat with brain regions that process reward information, but in some cases the subregions or neuron sub-populations that mediate these two processes are distinct even within the same general CNS region.  Opioid signaling in these regions is a major factor in hedonic processing, but has less impact on reward functions (5, 6).  Opioids, in certain parts of the brain, are pleasure chemicals.

Another class of chemical signals worth mentioning is the endocannabinoids.  The endocannabinoid receptor CB1 was identified as the brain receptor for delta-9-THC, the primary psychoactive constituent of marijuana, and then the endogenous ligands for the cannabinoid receptors were discovered (7).  We now know that the main two endogenous ligands are anandamide and 2-arachidonylglycerol, and the two receptors are CB1 and CB2.  Endocannabinoid receptors are expressed in CNS regions related to reward and hedonic processing, and they are known to be involved in both processes.

In addition to these three chemical systems (dopamine, opioids and endocannabinoids), there are other neurotransmitters that are intimately involved such as glutamate and GABA, but these are not  specific to reward and hedonic processes since they are the primary excitatory and inhibitory neurotransmitters of the brain, respectively.  So their involvement in reward and hedonic processing depends strictly on which neurons they're acting on.

Testing the Food Reward Hypothesis


 If the food reward hypothesis is correct, we should expect to observe certain things:

1. Increasing the reward/palatability value of the diet should cause fat gain in animals and humans.

2. Decreasing the reward/palatability of the diet should cause fat loss in animals and humans that carry excess fat.

3. Individual sensitivity to food reward should predict future fat gain.

4. Brain circuitry that controls motivational and hedonic processing should interact with circuits that control food intake and body fatness.

5. Manipulation of reward and hedonic circuits in the brain (e.g., by lesion, drugs or genetic manipulation) should impact food intake and body fatness.

6. Genetic differences that influence reward and/or hedonic circuits should correlate with differences in body fatness.

7. On a cultural level, obesity prevalence should track with changes in the reward/palatability value of prevailing diet patterns.


In the next post, I will explore whether or not the evidence is consistent with these predictions.

115 comments:

jack said...

Stephan,

I'm curious to see how you frame the evidence for this hypothesis because it most certainly doesn't apply to me. I have experimented with diet for many years and have found unquestionably that a high reward and high palatability diet led to 40 pounds of weight loss as long as I cut out the carbs which eliminated my feelings of hunger. (My diet also brought all of my lipid levels back into their healthy ranges so my doctors tell me to continue doing exactly what I am doing.) I have kept this weight off for years now while starting every day with delicious food like a bacon, egg and cheese omelette fried in butter. I have zero reason to eat anything that is not delicious and so I don't and I can also not imagine eating bland food ever again.

Clearly my own experience does not fit your hypothesis so I'm curious to see if you believe this theory is universal. I'm looking forward to part 2.

Anonymous said...

If William Davis ("Wheatbelly") is correct, then wheat does in fact manipulate those circuits, so Stephan's item 5 should be in play. If so, then your good results would not be related to bland vs. delicious, but to manipulated brain circuits instead.

Anonymous said...

jp - I think the short answer is that low carb diets are generally not high reward/high palatability diets.

Sure, bacon, eggs, cheese fried in butter sound tempting if you are viewing them as diet no-nos and thus high reward/palatability.

I eat all of those things AND carbs and lose weight too. I do calorie counting. That said, there are both non-carb'ish and carb'ish things that I avoid because I tend to over-indulge and then gain weight without tracking carefully.

jack said...

elo-sf, the short answer has nothing to do with generalities of low carb diets and everything to do with my experience of a high reward/high palatability diet that is completely counter to the hypothesis and how Stephan might account for such variances in the human person in his next post.

lost2ny said...

What in the world are you talking about: "bacon, eggs, cheese fried in butter sound tempting if you are viewing them as diet no-nos".

What? They are very tempting! They are some of the most delicious and rewarding foods known to humans! In 2004 I lost 65lbs on a low carb diet without counting a single calorie and eating a much larger and more satisfactory diet than I had for the previous 6 years when I was trying to calorie count (unsuccessfully, might add - every 10lbs I lost that way was followed by a 20lbs gain when I broke down and couldn't stick to it!).

I kept that weight off for roughly 6 years until I started eating carbs again a year and a half ago. Now I'm up 30 and it's time to get serious again. Modern life and food manufacturing make it far too easy to get WAY too many carbs every day.

That being said, I'm also very much looking forward to part 2, because there's something that just doesn't make sense about all of this. My own theory is very simple: humans did not evolve to eat large amounts of carbs and large amounts of protein at the same time. In fact, I wouldn't be surprised to find that switching back and forth between them every few months is the most healthy diet of all. Perhaps that's consistent with this hypothosis - we'll see.

Christopher said...

Jp - eggs and bacon meal is less rewarding than a doughnut, or likely even a bagel.

Was your previous diet really less rewarding than your current? What were you eating before that you don't now?

CelticPhoenix said...

Stephan, I think this whole food-reward concept needs to be connected to gut flora, biofilms, and intestinal health. For instance, you (and others) have pointed out the efficacy of “bland, liquid” diets for fat loss, and you even referenced in a previous post how already lean people tend to still consume quite a bit of calories while on the liquid diet, whereas overweight individuals consume much less and lose fat. You attribute this to palatability and neurological “reward” signaling (mediated partly by leptin).

However, I think an equally plausible theory is to consider gut flora and intestinal health. Liquid and “elemental” diets have long been used successfully to heal colitis and leaky gut. I have even read anecdotal reports about it not just alleviating symptoms temporally, but actually curing people, where they can begin eating wheat and grains and cakes again (for YEARS) without having further pains and problems. This tells me they are actually cured. In any case, obese people may be overweight, in part, because of inflammatory processes led by gastrointestinal distress and bad gut bacterial growth (see Art Ayers’ “Pre- and Probiotics” post on Cooling Inflammation). Diets such as the “Specific Carb Diet” (which would count as “bland” by you, I imagine) work wonders for autoimmune/gastro intestinal diseases, as well as, of course, losing fat. Antibiotics, alternatively, lead to initial fat loss and decreased markers of inflammation, but with long-term fat gain and increase in inflammation (both is humans and, of course, our Prime USDA Top Choice grain-fed cattle!). As Art Ayers from Cooling Inflammation points out, antibiotics often lead to increasingly complex (and harmful) gut biofilms.

In summary, I know Leptin and gut health are very connected (http://www.ncbi.nlm.nih.gov/pubmed/12007560), and I would like to see your thoughts on this and its connection to diseases like Ankylosing Spondylitis, Chron’s and even the new surge of interest in low-dose naltrexone (opiod antagonist). Two family members of mine were diagnosed with AS just this year, and I’d love to learn as much as possible (from smart people like you!) about the underlying mechanisms.

Angelyne said...
This comment has been removed by the author.
Christopher said...

Lost - you say you gained 30# when you started eating carbs again. What are your carb sources? Potatoes and white rice? Or bread, cookies, and cake?

Angelyne said...

I'm thinking that in this context high reward/high palatability does not simply mean tasty. Rather it describes food that makes you want to eat it sometimes past the point where you think you should have stopped. Like the slogan for those chips "Bet you can't eat just one". That doesn't apply for bacon and eggs, and indeed the majority of very low carb food.

Anonymous said...

I think the real challenge of these discussions is that EVERY diet plan (low carb, low calorie, low food reward) will tend to reduce to other items to a large degree as well.

It is VERY hard to have a high calorie, but low carb diet. And similarly it is hard to have a high food reward, but low carb diet.

Also, the question is not what any one of us finds personally very tempting--our trigger food so to speak. But rather, which of a set of choices would more folks in a group of say 100 people pick? Further which would tend to drive over consumption? Etc.

jack said...

Angelyne - "I'm thinking that in this context high reward/high palatability does not simply mean tasty. Rather it describes food that makes you want to eat it sometimes past the point where you think you should have stopped."

How is that definition not circular?

Tastiness is central by Stephan's definition: "Palatability refers specifically to the enjoyment derived from a food, also called its hedonic value."

Anonymous said...

McDonald's french fries have high food reward, but aren't that tasty. Most people want to finish the entire box of fries.

Bacon and eggs are tasty, but not as high in food reward. Many people can push away a partly eaten plate of eggs and bacon.

Chris Kresser said...

jpblogtest,

Your experience doesn't contradict the food reward hypothesis at all. You significantly reduced the reward value of your diet by cutting back an entire class of macronutrient (carbohydrate). As Stephan has explained before, each macronutrient has reward value and if you reduce one of them the reward value of the diet is lessened. This is one explanation for why both low-fat and low-carb diets can work in the short-term.

jack said...

Christopher - I disagree. We can learn to like foods and dislike them as well; we can acquire tastes and the opposite. This fact impacts the reward side of the hypothesis and the fact that I no longer have a taste for some nasty carb foods doesn't mean that my overall food reward has been lessened but rather it has been displaced entirely to other foods that I do like. Your reply also ignores the clear palatability benefit of my eating only foods that I love to eat which is clearly against the proposed theory.

And I am not alone. I know many people (personally, not online) that have had this same experience with food who now realize that they are free to eat and enjoy food again as long as they avoid carbs and they can lose weight if they wish because they are no longer hungry.

We (I don't know if it is everyone) can enjoy our food and lose weight at the same time by avoiding carbs. My only question for Stephan is how does he explain our experience with his theory if his theory indeed applies to all humans.

Chris Kresser said...

No. There is more to palatability than what you like. You need to go back and read this post: http://wholehealthsource.blogspot.com/2011/05/food-reward-dominant-factor-in-obesity_26.html

"Reward is a psychology term with a specific definition: 'a process that reinforces behavior' (1). Rewarding food is not the same thing as food that tastes good, although they often occur together."

"It doesn't matter whether or not you like the Little Debbie cake once it's in your mouth. It doesn't matter how you feel afterward. The only thing that matters is whether or not you'll buy another one tomorrow. That's food reward."

I've yet to see a critique of the food reward hypothesis from someone who actually understands it.

jack said...

Christopher - What I notice from your argument is that:

1) you keep ignoring the high palatability factor in my counter argument and

2) you use reward in a convenient way as:

a) you say that palatability and reward do not necessarily coincide yet you cite only an example of when they presumably do (e.g. little debbies)

b) if they do not coincide then you must argue that people gain weight because they feel compelled to eat food they find gross

Really? If you (& Stephan?) can ignore the palatability part and focus on reward then you are relying entirely on b? This must be the the case in the face of our counter-examples and that is one that will be challenging for Stephan to reconcile with our experiences.

Christopher said...

Jp- I think you may be confusing me and Chris Kresser in this thread. There is some other confusion building as well.

I too love bacon and eggs, but it's not important how I feel about the food (whether it be bacon or a doughnut), what is important (to this hypothesis) is the chemical environment created in my brain as a result of consuming the food.

Also, we all need to be careful about how we understand the mechanism elucidated in this hypothesis. Angelyne brings up lays potato chips and the "betcha can't have just one" slogan. This is an example of a rewarding food, imo.

It is my understanding that the hypothesis is not that anyone who eats potato chips will eat potato chips until they get fat because they can't stop themselves. It's that eating potato chips triggers a reward response, which raises the set point, causing your body to "defend a higher fat mass," perhaps with changes in appetite and metabolic rate, among other things.

In other words, you could raise your set point with potato chips and then get fat on anything - though consuming low-reward foods could lower your set point again.

Stephan touched on the fact that low-carb and low-fat diets cause similar weight-loss patterns - which to me is the nail in the coffin for the CHO, as I blogged about here: ( http://www.engrevo.com/blog/food-reward-vs-carbohydrate-hypotheses-of-obesity/ ) You can skip to the graph at the end (which I stole from Stephan) for why I think this is so important.

Christopher said...

jp - I think that bacon and eggs is a perfect example of palatability and reward not being coincidental.

Also, I wouldn't be surprised if the bottom line here is that aside from shangri-la type fat loss diets, the big takeaway here will be to avoid food that has been engineered to be hyper-palatable.

jack said...

Christopher - Yes, I have been confusing you with Chris. My bad.

And yes how you feel about the food is indeed half of the theory, the palatability part, and can't be so easily ignored. It is the reward/craving aspect that is in contention at the moment.

" I think that bacon and eggs is a perfect example of palatability and reward not being coincidental."

I find bacon and eggs to be extremely palatable and I eat them every day. They are not a good counter-example though because they are high protein/fat.

The fact remains that myself and others have lost much weight with highly palatable diets that reduce our cravings (reward) so lumping the two together as the food reward hypothesis does necessitates an explanation for this or else it doesn't apply to all of us.

Dregs said...

Stephen,

This was helpful to those of us who were feeling that, while intriguing, FRH was not being expressed in scientifically testable enough terms to get a firm grip on.

That said, some of the distinctions are not being made in a consistent enough way as I read them.

For instance, Stephen's expected observation #2 states that:

"Decreasing the reward/palatability of the diet should cause fat loss in animals and humans that carry excess fat."

I understand why we would expect high reward foods (creating desire and motivation to eat) to result in fat gain. I fail to see how high palatability / hedonic value foods would necessarily do so in the absence of high reward. I.e., a highly palatable food can bring great pleasure without triggering desire to eat more (unless it is ALSO a high reward food).

Would the concept of satiety be useful here? Satiety seems to be the flip side of reward: reward signals you to eat more, satiety signals for you to stop eating.

My lay understanding of the usefullness of "paleo" style eating is that it significantly reduces reward while still keep palatability very high. So the brain is given the pleasure of eating but with excellent satiety, so that there is little desire / motivation to overeat. As pleasure and satiety are both maximized, the diet is much easier to keep for many people.

On that basis, I don't understand Stephen's implication that reward and palatability are interchangeable in leading to obesity. Perhaps it is not more than his use of a slash (reward / palatability) to create an ambiguity in prediction #2 (does the slash between 'reward / palatability' mean "and" or "or"? Very big difference.)

@jpblogtest:

For the above reasons, I also don't understand how you could describe your successful diet as consisting of both high reward and high palatability.

That would imply that you were in a constant state of desire for more food and needed to exert control constantly to avoid eating too much.

Saying you eat a high reward diet is the equivalent of saying that you eat a diet that makes you feel hungry. It does seem intuitive to me that a successful weight loss diet will be one in which food reward (i.e. the motivation to eat) is not significantly reduced.

If for no other reason than aesthetic and social, I find it hard to accept that a healthy approach to food is one in which a person is constantly obsessing about food, fighting off desire to eat or spending an inordinate amount of time eating. And the necessary implication of saying you "eat a high reward diet" is that one or more of those situations describes your state of mind at all times, either that or you are misusing the term "high reward".

Dregs said...

Clarification: I was talking about prediction #1, i.e. that increased "reward / palatability" will lead to increased fat storage. But point is the same: Is that either-or or both-and?

lost2ny said...

To answer Christopher's question as to my carb sources: Actually, potatoes and rice (usually brown - it tastes better!) are the primary sources with bread in the form of wraps being the secondary source. I don't really have much of a sweet tooth and only rarely eat cakes or ice cream or candy on social occasions, with the exception of chocolate (and even there, I prefer things like Lindt Sprungli 85%)

In addition, I would like to add that I was eating a LOT more food when I was loosing weight in 2004 than I had during my dieting phases in the previous 6 years.

Sue said...

Lost, obviously you can eat more food when losing weight because you were bigger.
How much potatoes and rice did you eat if it caused 30 pounds weight gain over the years. Why change up your diet if you were maintaining for 6 years?

Sue said...

Lost, also if you added those carbs into your diet did you decrease intake of the other usual portion of your diet?

jack said...

Dregs - You are correct. With reward defined as craving then I have a high palatability/low reward diet and your post is an excellent summary of the likely conflation of palatability and reward in the FRH. You said it quite well.

I will now wait for part 2 to hear Stephan's reply.

Craig said...

I guess the confusion I have with the food reward concept is that I haven't seen it tied to anything that is independently measureable, and so it ends up feeling like a slippery, circular arguement:

People get fat because they over eat. They over eat because their diet has a high food reward value. What is food reward value? It's what makes you over eat, despite your best intentions! Case closed! But so far, I haven't learned very much.

Maybe that will be in part 2?

Stephan Guyenet said...

Hi jpblogtest,

Everyone is different, and I am not claiming reward/palatability is the #1 factor in all people. However, I don't think you have established that the concept does not apply to you. Did you ever try a low-palatability low-carbohydrate diet? It's possible that it would have caused more fat loss than a higher-palatability low-carbohydrate diet.

Bacon, eggs and cheese are palatable, but are they more palatable than cookies, Doritos and French fries? Or are they more palatable than the low-carb foods you mentioned foods plus a baked potato? I can't speak for you, but to most people they aren't.

Low-carb diets are reduced in reward value, particularly if they are centered around home-made foods which it sounds like yours is. Even if the diet is more delicious to you than one including those same foods plus carbohydrate, which I doubt, it is still a reduced reward diet.

That being said, it seems to be working well for you, and that's the whole point, so I'm not questioning that aspect of it or suggesting that you change course.

Hi Craig,

I don't understand the "circular argument" critique that I've heard from several people. Palatability is how much you like a food. Reward is how much it reinforces behavior. Neither of those are defined on the basis of how much they drive food intake. So how is it circular to say that they increase food intake?

Flare said...

In ob/ob mice, blocking or enhancing CB1 can cause an 88% decrease or 100% increase in plasma lipopolysaccharide, respectively, likely via modulation of gut barrier function. CB1 expression in mice can be decreased -25% by prebiotics, -60% by antibiotics, and increased 160% by a 'high fat' lab diet (but only by 60% if the HFD includes prebiotics). Altering mice gut microbiotas via prebiotics and probiotics can significantly modulate intestinal permeability

CB1 receptor knockout mice: protected from diet-induced obesity, despite similar caloric intakes as mice who do become obese.
CB1 blocker in diet-induced obese mice: -50% reduction in adiposity, correction of insulin resistance and lowered plasma leptin levels.
CB1 blocker in obese monkeys: -23% reduction in food intake, bodyfat by -39%, leptin by -34% (pair fed animals did not experience improvements).
CB1 blocker in humans: 4.7 kg greater weight loss over 1 year, compared to placebo.

Continuously injecting lipopolysaccharide (LPS) into mice, at levels that mimic the endotoxemia seen in metabolic-syndrome mice, causes glucose levels, insulin levels, and weight gain similar to 'high-fat' fed mice. Continuous LPS can cause insulin resistance in cats. A single LPS injection can cause a 100% increase in serum leptin levels and 44% increase in triglycerides. Continuously injecting humans with endotoxin can cause a 35% decline in insulin sensitivity, and increases adipose tissue inflammation. Inflammation within adipose tissue occurs during obesity, and interrupting this inflammation prevents metabolic abnormalities.

"In conclusion…we found that metabolic concentrations of plasma LPS are a sufficient molecular mechanism for triggering the high-fat diet–induced metabolic diseases obesity/diabetes."

Mice deficient in certain immune-fighting functions have altered gut flora and develop metabolic syndrome.

Transferring an obese mouse's gut flora to germ-free mice causes obesity in the colonized mice.

17% of morbidly obese have small intestinal bacterial overgrowth syndrome and circulating LPS (a type of endotoxin) is 76% higher in type 2 diabetics compared to controls. Herpes simplex 1 (a virus) concentrations correlate with percentage fat mass. HSV-2 is negatively associated with insulin sensitivity after controlling for BMI, age, and CRP . Genetically obese (leptin deficient) mice have enhanced intestinal permeability which leads to increased endotoxin levels. The microbiota of type 2 diabetic mice and obese mice and humans are significantly different from their lean counterparts.

"four viruses have been reported to induce obesity (infectoobesity) in animal models (chickens, mice, sheep, goat, dogs, rats and hamsters)"
"infection can be transmitted horizontally from an infected chicken to another chicken sharing the cage"
"The majority of children found to be AD36-positive were obese (15 [78%] of 19 children). AD36 positivity was significantly (P<.05) more frequent in obese children (15 [22%] of 67 children) than nonobese children (4 [7%] of 57 children)."

The CB1 receptor can significantly modulate drug and palatable food seeking behavior in mice, including sugar, chocolate, 'emotional behavior', anxiety, stress and depressive-like behavior. CB1 blocking reduces reward seeking in rats.

jack said...

Stephan,

Thank you for your reply. I've no reason to try a low-palatability low-carb diet because I now don't need to lose any weight and I like good tasting food vs. the alternative (!) but your introduction of that hypothetical as also likely to achieve the same goal reinforces the question as to why palatability is involved in this discussion at all if it is not a driver of reward/cravings/hunger.

I hope to hear (in part 2 perhaps) why low-palatability is better than high-palatability since it is a distinct variable. You must think that it is important to eat bland food to include this criteria so strongly in your hypothesis but I don't understand why.

Also, how are these things to be measured? Palatability is a personal preference and you can take a survey but how is reward/level of craving measured?

thanks.

Flare said...

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2890937
http://ajpendo.physiology.org/content/292/3/E740.long
http://www.nature.com/oby/journal/v15/n1/full/oby2007541a.html
http://care.diabetesjournals.org/content/29/5/1058.long
http://ajpgi.physiology.org/content/292/2/G518.long
http://www.ncbi.nlm.nih.gov/pmc/articles/pmid/21808634
http://www.ncbi.nlm.nih.gov/pubmed/16033867
http://www.ncbi.nlm.nih.gov/pubmed/21831779
http://www.ncbi.nlm.nih.gov/pubmed/21829158
http://www.ncbi.nlm.nih.gov/pubmed/19774074
http://www.ncbi.nlm.nih.gov/pubmed/20203013
http://www.ncbi.nlm.nih.gov/pubmed/20203013
http://www.ncbi.nlm.nih.gov/pubmed/20535027
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2925525
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2702831
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2675276
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1087161
http://www.ncbi.nlm.nih.gov/pubmed/16570103
http://ajpregu.physiology.org/content/284/2/R345.full
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2844840
http://www.ncbi.nlm.nih.gov/pubmed/20605903
http://www.ncbi.nlm.nih.gov/pubmed/18022033
http://diabetes.diabetesjournals.org/content/56/7/1761.long
http://www.ncbi.nlm.nih.gov/pubmed/21266508
http://www.ncbi.nlm.nih.gov/pubmed/15234252
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2797919
http://ajpendo.physiology.org/content/292/3/E740.long
http://www.ncbi.nlm.nih.gov/pubmed/21507694
http://www.discoverymedicine.com/Satoshi-Nishimura/2009/09/22/adipose-tissue-inflammation-in-obesity-and-metabolic-syndrome/
http://diabetes.diabetesjournals.org/content/56/7/1761.long
http://diabetes.diabetesjournals.org/content/56/7/1761.long
http://diabetes.diabetesjournals.org/content/56/8/1986.long
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2841839
http://www.fasebj.org/content/22/7/2416.long
http://diabetes.diabetesjournals.org/content/57/6/1470.long
http://www.ncbi.nlm.nih.gov/pubmed/18626197
http://www.ncbi.nlm.nih.gov/pubmed/20724524
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2922379
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2922379
http://www.ncbi.nlm.nih.gov/pubmed/19138827
http://www.ncbi.nlm.nih.gov/pubmed/17908526
http://www.ncbi.nlm.nih.gov/pubmed/11443497
http://www.ncbi.nlm.nih.gov/pubmed/20855385
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3083881
http://www.ncbi.nlm.nih.gov/pubmed/19660493
http://www.ncbi.nlm.nih.gov/pubmed/18448611
http://www.ncbi.nlm.nih.gov/pubmed/12897210
http://www.ncbi.nlm.nih.gov/pubmed/15083252
http://www.ncbi.nlm.nih.gov/pubmed/20381618
http://www.ncbi.nlm.nih.gov/pubmed/17392737
http://www.ncbi.nlm.nih.gov/pubmed/18469537
http://www.ncbi.nlm.nih.gov/pubmed/11823890
http://www.ncbi.nlm.nih.gov/pubmed/16250845
http://www.ncbi.nlm.nih.gov/pubmed/21471378
http://www.ncbi.nlm.nih.gov/pubmed/17728698

Unknown said...

food reward is not in question. Most people get the science and how well it is worked out. Since it is an outflow only tract act upon hypothalamic neurons it is not dominant to anything. Moreover reward as you described it in your last series is based upon taste. Leptin has receptors on the taste buds of all three cranial nerves of the tongue. Leptin acts upon the hypocretin neurons that are proximal and control all reward tract firing. This theory is a road to no where until you you discuss how the reward tract somehow modulate the hypothalamic leptin receptors. That has never been reported in the literature either. No one is disputing the reward tracts play a role. People are disputing your use of the word dominant with regards to obesity.

Dregs said...

@jpblogtest: Thanks. I tend to think of "reward" -- as Stephen uses the word -- as meaning "craving-inducing", not "craving" itself.

I do think that some of confusion around this topic comes from the fact that ordinary words are being given very technical definitions that don't entirely jibe with their customary usages.

That's fine -- every field has its so-called "terms of art" -- but I think eveyone needs to understand those definitions and MEAN them consistently when the words are used if we aren't to talk past one another.

Stephen,

In your response to jpblogtest, you asked a rhetorical question: "bacon, eggs and cheese are palatable, but are they more palatable than cookies?", with the implication that they are not, or they are not for all people.

Again, as I understand the two terms of art, I would say that the major distinction betweeen bacon and cookies is one of "reward", not palatability: We tend to overeat cookies but not bacon, not because the cookies taste good (palatability) but because they reinforce a desire to eat more ("reward").

Per my previous comment, I think it may be better to describe things in terms of "satiety": Bacon induces it, cookies do not. Saying something is more or less sating is understood by everyone; saying something is more or less "rewarding" will not be, and the confusion will continue...

Now that you have clearly defined the meanings of each of "reward" and "palatability", it seems like it would be best to avoid conflating them by writing "reward / palatability". The tendency to "overeat" comes not from something tasting good (palatability), but from its creating a desire to eat more of it (reward or satiety).

Anonymous said...

For anyone interested in further reading on the subjects of wanting, liking, hedonic impact, and incentive salience, I've been exploring them for some time as part of my ongoing series "Why Are We Hungry?"

Part VI goes into depth on the subject - though as "liking" and "wanting" are only two of the four hunger drives (plus willpower), the entire series may be of interest.

I look forward to seeing how Stephan integrates these concepts into his previous explanations and presentations.

JS

Wanna Be Mother said...

"There is rarely a single experiment, or single study, that can demonstrate that a hypothesis is correct."

There is actually no experiment that can demonstrate any hypothesis is correct. One fundamental thing about science is that all we can do is really try to disprove hypotheses, we can never prove anything from an experiment. True, when much evidence has been gathered and it is all consistent with a hypothesis, we start to get more confidence that our hypothesis may be "correct", but it could still be disproven by a future experiment. In fact, we can almost certainly say that eventually most of our hypotheses will be shown to only be correct under certain circumstances or with certain corrections, even if they seen correct under many experimental tests (ie, Newton's laws only work for big slow stuff). Our hypotheses necessarily involve simplification with a model that doesn't take into account all the details, thus they can't be all true all the time, especially in biology/medicine where there are so many factors to consider and even experiments with results "consistent with the hypothesis" have so much variation. You could end up having two or three or more important effects that contribute to the development of obesity but it is hard to control all of these in any experiment, so you could get some good results for your hypothesis, but you're missing the big picture.

Sadly, science can't prove anything :( That is only for mathematicians. This is a detail about science that I am anal about because you get these people who don't understand science saying "evolution has never been proven" or "evolution is just a theory". To that I say "evolution is just a theory, kinda like gravity".

David Moss said...

A couple of questions:

-What's gone wrong with our regulatory mechanisms for reward to lead us to obesity in this way? As you've noted before, increased fat mass, via leptin, ought, one would think, to decrease our interest in food. Or is this not the case and do we adopt a thrifty gene line of thinking, whereby if we come across certain kinds of food (not ubiquitous) we just eat as much as possible?

-Evolution doesn't seem to have been mentioned much since Food Reward became the hot topic. What's the point of our being wired this way. This is most especially trenchant in the context of your reward-raises-setpoint hypothesis. What's the point of that? One can imagine pretty big variations in food reward/palatability throughout our evolutionary context. Depending on whether we're faced with 50% fat meat or 10% fat meat our food will, presumably, have radically different reward values. Yet one might have thought that we'd ideally be programmed to eat our 3000kcal [or whatever] from the food available, not to eat radically different amounts depending on the rewardingness of the meat.
------


And FWIW here are my answers to Stephan's questions to jpblogtest. Hanging around the online health community for even a short while, you encounter loads of people with the same pattern:
"Did you ever try a low-palatability low-carbohydrate diet? It's possible that it would have caused more fat loss than a higher-palatability low-carbohydrate diet."
- Yes. For years (as a teen and young man) I struggled to maintain/lose (a very small amount of weight) on what I can only assume was a very low reward high carb food. I followed conventional health advice to the extreme. Ate masses of low calorie, high fibre foods, as raw/unprocessed as I could bear- lots of raw cabbage, steamed brocolli, tomato 'soup' that was just pureed tomatos. Steamed carrots were one of my more calorically dense foods. I ate canned beans after a workout and raw oats were considered a treat. I was basically constantly eating and constantly hungry and still didn't lose fat mass.

I did a bit better eating plain wheat germ and soy milk (which I found delicious and still think about longingly) while running 10k per day. After switching to eating low carb- cheese and avocado being staples- I instantly lost all appetite, had to remind myself to eat and for the first time in my life didn't feel hungry. Protein intake didn't increase and was, in fact, quite a bit lower than in my high carb days.

"Bacon, eggs and cheese are palatable, but are they more palatable than cookies, Doritos and French fries? Or are they more palatable than the low-carb foods you mentioned foods plus a baked potato?"

Palatability seems highly variable. In my HCLF days, a dessert spoon of cream tasted like elixir of the Gods. Very quickly after going LC, pure fat just tasted bland. Eating fat barely seemed like eating at all.

Plain potato was the staple of me and my family (all the rest of whom are obese), throughout my childhood, who all follow the British standard of meat, two veg and lots of potato. I always found starches very boring and unpalatable, but find them extremely hunger-generating. Basically, at every meal, eating plain potato or rice (and I do mean entirely plain- I shudder even to recollect) would stimulate my hunger, to not feel hungry again I could either eat more of other things (which since the available foods were carbs wouldn't do very much), or I could try to hold out for a few hours by sheer willpower until I reentered a fasted state and the hunger disappeared. Even now, adding a potato worth of carbs to my high fat, proteinous diet recreates hunger very quickly (I gained a stone in 6 months when I tried this). It would take a huge amount of non-carb food to displae this hunger.

Note that through all this I was never especially overweight, maybe a stone or so and was young, latterly very active etc.

David Moss said...

Oh and a couple of related points:

I think that there are at least a couple of different populations in play here (obviously with a continuum running between them). The food reward discussion seems most powerful in relation to people in the general population mindlessly and compulsively eating doritos, cookies, potato chips etc., including even when they're not hungry. Nevertheless, it seems far less well placed to deal with the innumerable people prevalent in the online health community who struggle to lose weight and indeed who fail to not gain weight, despite constantly starving themselves, miserably on (typically) low fat and, one assumes, rather unrewarding diets.
Similarly the advice to 'just eat a whole food diet of bland potatos' is probably apt advice for not becoming very obese, where the contrast class is a diet of soft drinks and doritos, but I suspect it'll be woefully inadequate for most individuals in industrialised coutries to avoid being significantly overweight. (Although maybe after adequate supplementation and a few generations of only eating potatos and the removal of all environmental toxins, we'll all be metabolically undamaged enough to subsist on starchy wholefoods.)

It seems to me that one of the big points of contention between the Food Reward 'safe starches are fine unless you're a diabetic' versus LC 'LC is desirable, unless you're a young, undamaged athlete' is working out how many people really fall into the high carb is safe/low carb is optimal camps.

RawHemp said...

Reading the constant barrage of comments from low carb advocates about how fat boiled potatoes made them makes me wonder how messed up these people could be. All these people either lie or have serious thyroid problems.

jehane said...

Stephan,

I've read your articles on food reward with interest, but have yet to see how they have any relevance for me.

I am a 400lb woman, and have been attempting to eat low carb paleo for a little over a year, I have lost 40lbs, but am experiencing a yo yo effect, from difficulties staying on plan.

The extreme cravings for processed and sweet foods always takes me off plan sooner or later.

I understand your theory behind food reward, and indeed when I eat the sweet and processed food, it doesn't taste that good to me, but undoubtedly sets off a chemical reaction in my brain. In fact after some foods I can actually feel the 'high' that the food brings.

Like all other 'diets' you seem to just be saying 'don't eat these things and you'll lose weight', which is all very well, and I'm sure some can manage it. But although you're suggesting a different mechanism as to how weight loss occurs, I can't see how your theory helps a morbidly obese person with these chemical reations occuring in the brain, as it seems to rely on willpower - ie the ability to ignore the cravings (or the desire for food reward, or addiction, or whatever you want to call it) and eat the low reward food.

Agreed, if I were in the experiment with the bland liquid to drink and no other access to food, I would have little interest in the liquid except to stop hunger and would undoubtedly lose weight - however, I live in the real world, with highly palatable food everywhere, and cravings that have nothing to do with hunger or palatability, but the chemicals released when the food is consumed.

Is this another 'this will work for you, but only if you have the willpower to resist' theories?

Is the need for 'food reward' a milder pre addictive state? Are there medications that can help people who have difficulties resisting the need for food reward - for instance fluoxetine, which is sometimes prescribed for binge eating?

I'm not trying to be difficult, but just trying to understand how this theory can help people like me?

Beth@WeightMaven said...

jehane, here are a couple of ideas you may find worth considering.

Re diet, I've found it essential to ensure that my normal diet is as nutrient dense as possible. I don't know for sure, but I suspect that my practice of eating either a small amount of liver or a couple of pastured eggs every day along with some other healthy fats (my faves are coconut butter, pastured ghee, and very dark chocolate) largely keep my cravings to a manageable level.

But triggers still happen, so to deal with the minor variety, I do a regular meditation practice (my version is assisted ... I think the trick is just to learn to easily switch back to parasympathetic mode). Between this and my diet, I find that when I get an urge to eat, I can actually sit with it.

Re the bigger cravings, well, the first thing you have to be aware of is that giving into them reinforces them. You're essentially learning that food equals stress relief. The series Your brain on porn is a great resource for understanding this (turns out desires for sex and food have lots in common!).

If you give in, then realize that the best time to get back on track is immediately afterwards. If you do the "WTF" thing and start stringing multiple cheat meals together, all you are doing is more reinforcing and it just becomes harder.

Hope this helps. I do not believe that it's essential that people like us give up hyperpalatable foods forever, but instead, that we treat them as potentially problematic. Less is more!

Olga said...

http://www.sciencedirect.com/science/article/pii/S0940299310000862

jehane said...

Beth,

I appreciate you taking the time to respond, however, I do eat a nutrient dense diet and advice to 'ignore the cravings' or 'get straight back on plan' is nothing new to me, and I do try to do this. However the cravings do not go away, constant willpower is required for me to eat only low reward food - this is highy difficult long term, which is my point, that for those with real difficulties with high reward food, it seems rather simplistic to say 'just eat low reward food'.

Compulsive and binge eaters experience a drop in seretonin when abstaining from the high reward foods, which in itself makes it very difficult to continue to abstain.

I really appreciate the fact that Stephan is demonstrating that there are complex neurochemical reactions involved in eating foods high in reward, but for those of us where eating high reward foods is bordering on (or is) compulsion or addiction, what can be done to help?

Or is this just a theory to enable those with mild to moderate obesity to lose weight, or slim people maintain weight? Or can it be expanded to be more inclusive of those with severe obesity and more severe issues around food reward?

Beth@WeightMaven said...

jehane, just to be clear, I know exactly of what you speak. I fell off the food wagon a year and a half ago over the Easter holidays and didn't stop until I'd gained all the weight back that I'd lost ... and my high weight was very close to yours. I practically required an intervention to get back on track.

My point is that as long as you give into the cravings you are reinforcing them. I think that getting off these foods is necessary for sufficient time to down-regulate the parts of the brain that are responsible. And yes, it's very, very hard.

If you think serotonin is an issue for you, perhaps you might want to consider DesMaisons' Potatoes, Not Prozac a try?

Alternatively, Nora Gedgaudas has a number of suggestions for supplements to try and ward off cravings.

I'm not sure what your financial situation is, but what I also did from Oct thru April of last year was ~60 sessions of neurofeedback. That was a *huge* help, but alas, it's not covered for eating-related issues.

I'm not sure which of any of the things I've done have freed me from perpetual craving, but I am no longer suffering from it.

Drew Stearns said...

I find the Food Reward / Carbohydrate Hypothesis debate interesting, but they both effectively ignore the impact of physical activity.

If you take an unromantic, dispassionate look at food as fuel, things become much more clear. Lyle McDnold is a good example of this kind of thinking.

If you want to get more orthorexic about it, consult Rob Wolff or Kurt Harris.

Food isn't fashion.

Mike said...

I almost wonder if the treatment of obesity needs to be treated more like recovering from an addiction. somehthing similar to rehab might even be plausable.

I am curious though what some of the other commenters have mentioned about the gut flora, and also about how that affects your neurochemicals. doesnt a large amount of serotonin come from your digestive system?

I'm also curious if the corelation between depression and obesity is relevant here as well. Since things like dopamine and addictiveness seem to coincide here.

I'm sure I've said something horribly incorrect. Hopefully you can correct me instead of being offended.

ps. to those talking about dealing with cravings while in weight loss mode, I have heard that schedulling a cheat day, maybe once a week or so can help, you will still be on a downward trend, and being able to hold out just one more day is a lot easier than telling yourself "i'll never get to have that ever again". it might be something to look into.

Sam Knox said...

@Chris Kresser

"It doesn't matter whether or not you like the Little Debbie cake once it's in your mouth. It doesn't matter how you feel afterward. The only thing that matters is whether or not you'll buy another one tomorrow. That's food reward."

I'm not sure how much Stephan hangs his hat on this single statement, but it seems, on it's face, absurd.

If I buy a Little Debbie cake, or any other food, take one bite and spit it out in disgust, it seems unlikely in the extreme that I'd buy another one the next day.

It implies that a food can not only taste bad and have no discernible positive effect and yet, somehow, create the desire to consume it again.

jehane said...

Beth,

Again, thank you for taking the time to respond, I do appreciate it.

I am already currently looking into L tryptophan or 5 HTP, to help with the cravings and bingeing, rather than going straight for the prozac to boost the seretonin levels.

For me, the food reward, or addiction issue where certain foods are concerned is the last piece of the puzzle, and the hardest one for me to deal with.

I had hoped to gain some insight from Stephan's food reward series, but there seems to be nothing on support (ie supplements, medication etc) for those people who have extreme difficulty with cravings or bingeing. As that seems to me to be very much tied up with food reward factor and neurochemical reactions to food.


Sam Knox,

The way I see it (and I may be wrong), is that susceptibility to food reward factors is on a spectrum - for some people, the reward factors of foods hardly affects them at all, they could eat one bite of cake and have no desire to eat more, they can easily make food choices based on nutritional value, only eat if hungry, always stop when just satiated, and never obsess about any foods.

Then you have the other end of the spectrum where hunger or satiation has no bearing on the urge to eat - this person will eat when full because they feel compelled to eat the rewarding food - the food doesn't even taste that good to them when they eat it, but by this point they're not eating it for the taste but unknowingly for the release of chemicals it brings. they cannot understand how anyone would only eat half a chocolate bar and not need to eat the rest. No they are not greedy, but their brain reacts to certain foods like an addictive drug, rather than just as food.

I may well be way off, but that's how I see it.

Dragan said...

I enjoyed this post, as I did most in the series. That said, by your definition astronomy wouldn't be a science (we can't smash stars yet). Neither would computer science or other design sciences. Basically, it's a sloppy definition which detracts from the quality of your post, so I'd suggest minor revisions.

psychic24 said...

Only critique I have is in regards to Stephan's statement about not understand the circular logic, specifically:"Neither of those are defined on the basis of how much they drive food intake."

From what I understand, you keep stating that dense sources of calories, as well as macronutrient combinations (carbs+fat) within those dense sources increase food reward (and most likely palatability)--this means that as soon as something provides a source of increased food intake (in a convenient source, such as french fries or cake,etc), this will increase the food reward and people will eat it (over and over again).

Am I totally wrong?

Beth@WeightMaven said...

jehane, well, Stephan is a neurobiologist/PhD who is really looking primarily at the possible mechanisms (and look how many people are challenging him on that!).

Did you see his recommendations from a few months back?

You're right to presume that to some extent, these will work better for folks who aren't outliers on the overeating continuum. Michael Prager (Fat Boy, Thin Man) said that whether or not you believe in the concept of food addiction, he found that using addiction treatment (in his case, abstinence and spiritual support a la 12 steps) helped him resolve his weight from compulsive overeating.

I am with Porthos that for some folks, it may well be useful to be able to detox off SAD foods.

Mike said...

I am not disputing the food reward hypothesis. It seems plausible. I am also not a low carb fanatic. However, it seems the food reward hypothesis is a logical tautology. If a diet causes obesity, then it is causing you to eat more calories than you need and so it is automatically high in reward?

If that's the case, is there any way to possibly disprove the food reward hypothesis? Can the proponents of the food reward hypothesis design a study that might be able to disprove it? E.g., if you take a formerly obese person who has gotten down to a healthy weight following Kurt Harris' archevore diet (for example) and then that person adds in homemade potato chips deep fried in grass fed tallow and dipped in pastured sour cream (maybe with some herbs/spices added as in french onion dip) and allows himself to eat as much of these as he'd like.

If the cause of the obesity was the NADs (excess fructose, excess polyunsatured fats, gluten grains) then adding in these potato chips should not cause weight gain even if the person is allowed to eat as much as they want. However, if the cause of the obesity was food reward, then adding in these chips should cause weight gain. Would this be enough to disprove food reward hypothesis? I believe it's hard to argue these chips aren't high in reward for most people. But, if food reward hypothesis is a tautology, you could always argue that food reward is not disproven because the diet is still cutting out many other rewarding foods and we know this because the person is not obese.

luckybastard said...

@mike

that's an interesting point. i lost my weight and came down to be normal sized on a paleo diet that is identical to the archevore diet. once at a normal weight, three- sometimes more- times a week- i'd fry several potatoes mixed with sweet potatoes in bacon fat and put kosher salt, old bay, garlic and parsley on them and go to town. i didn't gain any weight. in fact, i continued to slowly drop weight. and i have always chalked that up to healing my metabolism by getting the NAD's out of my diet while going low carb for several months. Paul Jaminet suggested to me that eating kimchi and IF'ing while power-lifting also did wonders for my toleration of starch, which has always been a binge-trigger for me.

i still don't think that it discredits food reward, though. paleo and ancestral diets are naturally much lower reward- i would say on par with the evolutionary norm for reward- than the SAD. i think many of us here are eating in a whole foods fashion so i really don't think that alot of the food reward stuff applies, unless you're still having issues getting the weight off. i've done some n=1 with food reward(and i know a couple of guys who've done it even more hardcore, hopefully they'll speak up) and it does work to push down the setpoint if i go hardcore not seasoning anything. when i did go back to a normal paleo diet- all seasonings and variety back in- i didn't gain the weight back. i don't know exactly what that means about reward vs pallatability but i thought i'd share my experience.

bjk said...

For many people on a "paleo diet", they can easily forego a meal and start a fast without feeling sick. This jives with your "low reward" concept with it being so easy to go without.

Since there are profound differences in the palatability of foods among various cultures, I wonder how much of this is culture dependent.

This has finally clicked with me.

Thanks!

Rap said...

Despite repeated efforts by Stephan and other readers to clarify things, there seems to be continued confusion regarding the distinction between food palatability and food reward. But this isn’t surprising. Radical (Skinnerian-type) behaviorists much prefer the term “reinforcer”, rather than reward, when referring to consequences that strengthen (drive) behavior. They recognized early on that the common sense notion that rewards are pleasurable easily leads to confusion and a false assumption that consequences that we perceive pleasurable will necessarily strengthen the behaviors that they follow.

Might I suggest then that one way past this difficulty is to drop the distinction between food palatability and food reward. This shouldn’t be a problem from Stephan’s perspective since he is constantly linking reward and palatability together anyway. Moreover, the following article argues that the distinction between ‘liking’ (palatability) and ‘wanting’ (reward) is largely unnecessary.
---------
Appetite. 2011 Aug;57(1):286-94. Epub 2011 May 23.
"You Say it's Liking, I Say it's Wanting …". On the difficulty of disentangling food reward in man. By Havermans RC
Abstract: According to the influential theory of Berridge (1996, 2009), food reward comprises two components: food 'liking' and 'wanting', with 'liking' referring to the pleasure derived from eating a given food and 'wanting' referring to appetitive motivation. Animal research shows that these two components have separate neural correlates. In examining reward driven eating in man, researchers have thus begun to develop interest in indicators of 'liking' and 'wanting'. But validating 'liking' and 'wanting' requires the dissociation of these components when theory dictates they should diverge. One such circumstance is neural sensitization as this leads to exaggerated 'wanting' without increased 'liking'. However, there are no data suggesting that such sensitization is the likely result of (over)eating. Without sensitization, one cannot determine whether task performance is indicative of true food 'liking' or 'wanting'. It is concluded that it is important to assess appetite and palatability in the study of reward driven eating, but determining whether these measurements reflect either food 'wanting' or food 'liking' is not.
-------------------------------

So why not instead simply distinguish between homeostatic eating, driven largely by a biological state of caloric deprivation, and hedonic eating, driven largely by pleasure (or by reduction of “pain”—as when people eat to reduce stress)? The food reward hypothesis can then be viewed quite simply as “hedonic overeating” (eating that occurs in the absence of a homeostatic need for food). Whether a food is tasty or not then becomes largely irrelevant, apart from the fact that some types of tasty foods strongly encourage hedonic eating. And a major factor involved in that is the extent to which we are surrounded by very tasty, easily available, high calorie food items that readily elicit hedonic overeating.

Todd said...

Stephan,

I have the same objection to your hypothesis as Craig, jblogfest and others: it is either wrong or it is circular.

You say: "I don't understand the 'circular argument' critique that I've heard from several people. Palatability is how much you like a food. Reward is how much it reinforces behavior. Neither of those are defined on the basis of how much they drive food intake. So how is it circular to say that they increase food intake?"

If palatability is defined as reward or reinforcement of behavior, it can only be measured objectively by whether a person actually tends to eat more of it (relative to some baseline). So the Food Reward Hypothesis of Obesity"" reduces to: "Obesity is caused by people eating more of foods which they tend to eat more of". This is vacuous and without predictive content. It cannot be falsified, so it is not an empirical theory.

In the above definition of palatability, notice that palatability is defined in terms of the behavior only, not anything objective about the food.

Now, if palatability is defined as an inherent property of the food associated with "liking" or "wanting" -- rather than behavior -- then the Food Reward hypothesis must now hold that certain foods are inherently rewarding or reinforcing -- prior to any observations about actual behavior. in this case, the theory is empirical and testable. But in this case it is false. Foods are not inherently palatable. Some foods palatable in one culture are not palatable in others, and palatability to an individual can change radically over time.

It is more likely, I think, that metabolic changes (like insulin and leptin resistance in the brain) and psychological changes (like experience and acculturation) drive palatability, eating drive and obesity. Palatability is thus not a cause of obesity, but more often either a consequence or a correlative response to other changes.

I've explored this critique in more depth in my blog post on "Does tasty food make us fat?" (http://bit.ly/q1HWOL)

Todd

Stephan Guyenet said...

Hi Rap,

What you said makes sense.

Hi Todd,

The reasoning is not circular, because as I said before, the terms "reward" and "palatability" are not defined on the basis of how much of a food you eat. I will be dealing with that critique in a bit more detail in the next post, but here is a sneak preview:

Before diving in, I'd like to address the critique that the food reward concept is a tautology or relies on circular reasoning (or is not testable/falsifiable). I have never understood the basis for this critique. Neither the concept of food reward nor palatability are defined by their effects on food intake or body fatness. In the research setting, food reward is measured by the ability of food to reinforce or motivate behavior (for a description of a quantification method, see http://www.ncbi.nlm.nih.gov/pubmed/8879418). In humans, palatability is measured by having a person taste a food and rate its pleasantness in a standardized, quantifiable manner, or sometimes by looking at brain activity by fMRI or related techniques (http://www.ncbi.nlm.nih.gov/pubmed/14513063). In rodents, it is measured by observing stereotyped facial responses to palatable and unpalatable foods, which are similar to those seen in human infants. It is not a tautology or circular reasoning to say that the reinforcing value or pleasantness of food influences food intake and body fatness. These are quantifiable concepts and their relationship with food intake and body fatness can be, and already has been, tested many times in a controlled manner.

Anonymous said...

@Mike

"E.g., if you take a formerly obese person who has gotten down to a healthy weight following Kurt Harris' archevore diet (for example) and then that person adds in homemade potato chips deep fried in grass fed tallow and dipped in pastured sour cream (maybe with some herbs/spices added as in french onion dip) and allows himself to eat as much of these as he'd like."

I have seen people fail to lose fat or gain it back with just such maneuvers. They manipulate "paleo whole foods" into their own homemade versions of high reward food, and it subverts the effects of eating whole foods, suggesting that FR may be independent of other NADs, or even fairly considered a NAD itself - a cultural agent instead of a biological one affecting food intake. Susceptibility to this varies by individual perhaps, just as susceptibility to the SAD varies.

I see two confusions:

Palatability is a component of, but not identical to reward. High FR does not equate to "delicious".

Making the "circular" argument that FR is not falsifiable assumes that total caloric intake or fat gain are the only ways that reward value can be assessed, but I think that is not true. One could simply assess the likelihood that more food is sought or eaten in response to a given food than some other food, no? Why is the endpoint of obesity or fat gain required?

What is unfalsifiable about saying that the content of this bowl of chips results in the URGE to eat more of them subsequently than some other bowl of chips?

(And of course one might eat more of a certain chip even if it scored a 6 versus the other chip which scored a 7 when rank ordered by palatability - merely which one tastes better. Palatability is nor reward)

I think the mistake is in thinking that the final end point of obesity or actual fat gain is the only way to assess the presence of a FR effect. But by ignoring intermediate steps in the process (such as motivational behavior that occurs PRIOR to fat gain) one could say exactly the same about ANY putative cause of obesity or fat gain. That it is obesogenic if it makes you fat, and not if it doesn't.

But we know that is a nonsensical criticism.

Asim said...

Isn't food reward and pleasure sometimes dictated by the need of an organism for a particular vitamin or mineral, i.e. wouldn't the brain create a 'reward' value based upon the fact it needs, say, vitamin A?

Asim said...

Nothing taste better than the first sip of orange juice when you've been sweating outside in the heat, playing basketball for a couple of hours...

Beth@WeightMaven said...

Kurt writes:

"I think the mistake is in thinking that the final end point of obesity or actual fat gain is the only way to assess the presence of a FR effect."

I find it plausible that FR could play an important role in both bulemia and binge-eating disorder, both of which can occur in normal weight people.

Todd said...

@Kurt, @Stephan

Kurt: “Palatability is a component of, but not identical to reward. High FR does not equate to "delicious".

OK, then let’s focus on food reward itself.

Kurt: “Making the 'circular' argument that FR is not falsifiable assumes that total caloric intake or fat gain are the only ways that reward value can be assessed, but I think that is not true. One could simply assess the likelihood that more food is sought or eaten in response to a given food than some other food, no? Why is the endpoint of obesity or fat gain required?”
If you are right, then all one needs to complete the circle is the postulate that “increasing calorie intake (over a previous baseline) tends to result in fat gain”. That’s not a hugely controversial postulate.

To lay out the full argument:
1. Rewarding food reinforces (increases the likelihood of) increased food intake (versus some baseline)
2. Increased food intake results in fat gain.
Therefore,
3. Rewarding food increases the likelihood of fat gain.

This argument is unfalsifiable, so the reward hypothesis is circular.
What then about “palatability”?:

Kurt: “I think the mistake is in thinking that the final end point of obesity or actual fat gain is the only way to assess the presence of a FR effect. But by ignoring intermediate steps in the process (such as motivational behavior that occurs PRIOR to fat gain) one could say exactly the same about ANY putative cause of obesity or fat gain. That it is obesogenic if it makes you fat, and not if it doesn't.”

How does one quantify the “intermediate” motivational behavior independently of (and prior to) actual eating behavior? Perhaps you could survey people to get their subjective ratings of “palatability”, which according to Stephan has the dimensions of “liking” or “wanting”. Yet if palatability is not measured by behavior, it must be an inherent property of a food, or a relationship between the food and a person’s brain/mind. This leads to my other two objections, which nobody has yet responded to:

(1)”Palatability” cannot be an inherent property of foods, because it varies across cultures and individuals. Thus, it cannot be a reliable cause of overeating or obesity;

(2) If “palatability” represents a relationship between a food and a person’s brain/mind, that that does not mean that it drives eating behavior or obesity. The causation may in fact be the reverse: obesity and frequent exposure or conditioning could lead to a food becoming more rewarding. In fact, there is evidence that metabolic dysfunction (insulin and leptin resistance) interferes with the transport of insulin and leptin across the blood-brain barrier, interfering with the satiety function of the hypothalamus. Futhermore, the production of digestive hormones and neuropeptides can be conditioned, based on repeated exposure and expectation. So “palatability” may in fact follow, rather than precede, changes in metabolism and behavior.

Todd

John said...

Todd,

So obese have downregulated D2Rs I think. What if you can show D2R downregulation precedes/leads to behavior reinforcement leads to obesity? Does it matter if the foods that do so may be different for each person? This way there actually is more to measuring reward than behavior...? Now what if this ends up going along with the insulin hypothesis?

Dragan,

Many people don't think of computer science or certain physics branches as traditional, hard science.

jehane,

I think you're stuck having to use willpower, though it's tricky. Stephan has mentioned his old tendency to gorge on a box of donuts only to now have no problem avoiding them. It's a slow, gradual process. I have had the same experience as Stephan, and it's interesting to think about that [psychological extinction] in combination with food reward. I feel no stress in avoiding "junk" and haven't eaten it in a year--I wonder how my pleasure and reward would differ [if I did eat cookies] from the typical person binging on cookies after a bad work day.

Todd said...

@John,

Yes, the obese do have downregulated dopamine receptors. Gene-Jack Wang and Nora Volkow of the U.S. Department of Energy’s Brookhaven National Laboratory used PET scans to show that the obese have fewer dopamine receptors than normal-weight subjects. And the more obese the individual, the lower the number of receptors. Similar downregulation occurs in the brains of addicts and depressed patients, as I’ve detailed in my post at http://gettingstronger.org/2010/10/change-your-setpoint/

It is quite likely, as you suggest, that the trigger foods are different for each individual. But I think we have to ask why the intensified cravings or “hyperpalatability” are more pronounced in the obese. My suggestion is that hyper-responsiveness is a result of changes in the brains of the obese – receptor regulation and insulin resistance in the hypothalamus – not something that resides in the food itself. I would agree that eating lots of "junk food" often reduces the sensitivity of the hypothalamus to insulin and neuropeptides, which produces the phenomenon of hyperpalatability. But that is due to metabolism-altering properties of junk food, rather than it's "palatable" flavor. This contrasts with Stephan’s thesis, which reverses the causal sequence. Of course, once the brain is changed and foods become hyperpalatable, a viscious cycle of overeating sets in. So while I may agree with Stephan that hyperpalatability sustains obesity, I disagree that palatability is where things get started. Otherwise, all individuals would be equally susceptible to tasty foods – which is manifestly not the case. The changes have to get started in the brain.

dim sum massacre said...

I think the big problem with this discussion (and with Stephan's conception of the food reward theory) is that it singles out palatability as the dominant factor in determining food reward, while leaving out other factors that are just as important.

Satiety (which J Stanton written some interesting stuff about) is clearly one of those factors.

Consider the difference between potato chips and fois gras. I think most people would say that in terms of palatability, fois gras > potato chips. However, in terms of food reward, it's the other way around.

Why is this? it's because fois gras is highly satiating, while potoato chips actually tend to lessen sateity. After eating a single potato chip, one's desire to eat then next chip is increased. On the other hand, one's first bite tastes of fois tastes amazing, but with each subsequent bite the desire to eat more decreases. Thus fois fras is very hard to overeat, and potato chips are very hard NOT to overeat.


This seems to me to be a neat explanation for why the French are less obese than Americans: even though their cuisine is tastier (and thus more palatable) than processed American food, it's also much more satiating, and thus lower in food reward.

Anonymous said...

@Todd

"OK, then let’s focus on food reward itself."

We are.


"If you are right, then all one needs to complete the circle is the postulate that “increasing calorie intake (over a previous baseline) tends to result in fat gain”. That’s not a hugely controversial postulate."

You are talking in circles I think. That same statement is true for any cause of obesity and we don't dismiss all of them as as trivial.

"This argument is unfalsifiable, so the reward hypothesis is circular."

NO IT IS NOT. It is no more unfalsifiable than any other hypothesis about what makes us fat

We can compare different foods for their effect on the setpoint. One regime can be compared to another. They could have the same nutritive value and macro ratios and one may only differ from the other on how much is stimulates subesquent intake.

Your assertion of unfalsifiability is not sound.

You are confusing the difficulty of separating FR from other aspects of nutrition with the notion that it is completely unknowable.

"How does one quantify the “intermediate” motivational behavior independently of (and prior to) actual eating behavior?

I said independent of fat gain or caloric intake -yet can you not imagine an experiment that measures intake over the very short term or even assesses the desire to eat more without actual consumption or fat gain?

"Yet if palatability is not measured by behavior"

The behavior could be as simple and repeatable as the pepsi challenge.

And palatability could still be an important and even necessary component of FR without being the same as FR.

Torque and horsepower are parts of motorcycle performance that affect speed, but they are not speed and they are not even measured the same way.

With 0 torque there is no speed. Yet torque is not speed.

Generally more torque is required for more speed.

Yet an 800 lb harley davidson that has more torque than my 375 lb ducati is nowhere near as speedy as my ducati.

Because speed requires torque.

But speed is not torque.

Get it yet?

"(1)”Palatability” cannot be an inherent property of foods, because it varies across cultures and individuals. Thus, it cannot be a reliable cause of overeating or obesity;"

Non sequitur. Variability does not disqualify it at all, even if it is hard to measure. Would you say candidate preferences are too variable to affect election results? Palatability can be variable and learned and still be real and based on physiology as well as culture. We are not robots.

"(2) If “palatability” represents a relationship between a food and a person’s brain/mind, that that does not mean that it drives eating behavior or obesity. The causation may in fact be the reverse: obesity and frequent exposure or conditioning could lead to a food becoming more rewarding...... "

Testable hypotheses, yes.

Arguments that there can be no such thing as FR, no.

Todd said...

Kurt,

I think the reason we seem to be talking past each other (when we actually agree on many points) is that Stephan has used the term “food reward” to mean two very different things. I will quote directly from Stephan to support each of the two divergent definitions:

Behavioral food reward: “It doesn’t matter whether or not you like the Little Debbie cake once it’s in your mouth. It doesn’t matter how you feel afterward. The only thing that matters is whether or not you’ll buy another one tomorrow. That’s food reward.”

Biochemical food reward. Feeling satisfied after eating something is not reward. If you keep eating a starch food beyond what’s appropriate, that is probably because it has too much reward/hedonic value for you. Opioid signaling, implicated in hedonic processing, shuts off satiation signals in the brain and may also increase the setpoint. Dopamine signaling, implicated in reward, can strongly influence food intake and also seems to be able to increase the set point.

The problem is that Stephan vacillates between the two definitions, and they are quite distinct concepts. Behavioral food reward is merely the likelihood and degree to which a given food is consumed. Used in this sense, the food reward hypothesis is circular, because it holds that “foods more likely to be consumed will be consumed to a greater extent”. As you say, that statement is consistent with all theories of obesity. It is true, but it doesn’t tell us very much. I’m not sure I could think of an experiment that would falsify it – can you?

It is the second sense of food reward--biochemical food reward---that is more interesting and testable. I believe this is what Stephan intends by his term “palatability”, although I’m not sure. Chemical food reward involves hedonic processes, dopamine signaling, etc. But it is not necessarily the same concept as behavioral food reward. (Just as "torque" and "speed" are different but related concepts, to use your analogy). A food could be both behaviorally and chemically rewarding (palatable) or a certain person, but it could be one without being the other. For example, a food might cause a dopamine spike and yet fail to lead to overeating.
The food reward hypothesis is an interesting, testable hypothesis if “food reward” is understood in this second, biochemical sense – as palatability or hyperpalatability. But then we can’t go back and at the same time define food reward in terms of reinforcement and repeat behavior. It becomes an empirical question as to whether the palatability of food results in overconsumption, reinforcement and repeat behavior. We need to show that biochemically rewarding food is behaviorally rewarding – we can’t just define it to be that way....

Todd said...

(continued from above):

...If reward = palatability, then Stephan’s argument is that the obesity results from eating food that is palatable or hyperpalatable. My point about the variability of palatability is not that palatability isn’t real, but that it is not an inherent property of foods in themselves. If potato chips were inherently palatable and anyone eating them would head down the road to obesity, that would prove Stephan’s theory. But because some people find them palatable to the point of irresistibility, and some (like me) do not, that means palatability can’t explain obesity. It might not be a sufficient condition for obesity, and it might not even be a necessary condition for obesity. So how can it be a cause of obesity?

Studies that show obese people or animals eat a lot of palatable food don’t thereby establish palatability as the instigating cause of obesity, the spark that gets things going. They show only correlation, not cause. I will agree that once established, palatability might reinforce or sustain overeating, like adding fuel to the fire. But I think that this only works if your brain has already been changed, due to properties of the food, or your lifestyle, or genetics, than change your metabolism and your hypothalamus to make you susceptible to the signal of palatability. By itself, a given flavor doesn’t provide this motive force. It is plausible that foods become hyperpalatable due to processes which originate independently of their flavor.

Todd

Anonymous said...

@Todd

"Behavioral food reward is merely the likelihood and degree to which a given food is consumed."

No.

Availability, cost, expense, ease of preparation all affect "likelihood and degree'".

FR is the likelihood, given consumption of a food, that you will show behavioral motivation to eat more, whether you actually are allowed to or have the money to or if there is enough of it, etc..

It is not assessed by a food survey, anymore than top speed of a car is assessed by using a radar gun in a school zone.

The rest of your post so conflates palatability with reward that to respond would blow out the html limit.

It is really not that hard to understand, unless you have some motivation for not understanding it or not wanting to believe it.

Craig said...

@Stephan:

A lot of comments have since been addressed to the the circularity claim, and I am still digesting those. But this statement by you serves to illustrate my confusion:

"Reward is how much it reinforces behavior. Neither of those are defined on the basis of how much they drive food intake. So how is it circular to say that they increase food intake?"

But isn't food consumption the behavior being reinforced? If food intake isn't the behavior being reinforced, then what is the behavior that is being reinforced by the food reward.

Steve_Roberts said...

Todd:
I have an "N=1" to share.
I have recently encountered the cinnamon bun. Despite eating LC generally, I had one with morning coffee one day. It tasted great, but I didn´t feel so good a couple of hours later - all that gluten and syugar, maybe. However, the next day at coffee time I definitely felt a temptation to have a cinammon bun again.

Now if the above could be put on a scientific basis and extended to a large N, and beyond a single daily bun to the point where some effect on obesity is shown, then Stephan wins, in the sense that FRH really is important in obesity. I certainly don't dismiss the possibility.

Please note also, there is not a strong contradiction between FRH and the efficacy of LC / Paleo eating. Most LC / Paleo eaters will be avoiding the foods which have been engineered to trigger Food Reward, and will not be subject to their effects.

Coach said...

Dr. Harris,

"Torque and horsepower are parts of motorcycle performance that affect speed, but they are not speed and they are not even measured the same way.

With 0 torque there is no speed. Yet torque is not speed.

Generally more torque is required for more speed.

Yet an 800 lb harley davidson that has more torque than my 375 lb ducati is nowhere near as speedy as my ducati.

Because speed requires torque.

But speed is not torque.

Get it yet?"

Here's my beef with you - for years I have been reading your analogies which make little representation of the original idea in question... and then you subsequently talk down to someone, such as, "Get it yet?", when you haven't even explained yourself in the analogy, yet think you have. This is the drive behind my calling you a narcissist. You just don't have the right to do this, as much as you might think that you do.

Torque speaks to acceleration, not speed (by which I suppose you mean velocity). Horsepower speaks to velocity after torque has gotten you there.

That the 800 lb Harley that can't accelerate as quickly as your Ducati doesn't mean it has less (or no) torque, just that it doesn't have enough torque to displace its weight like the lighter bike can.

I still appreciate your work and intelligence but when you get outside of medicine for analogies, you have historically fell on your face, Sir. And that wouldn't be awful in and of itself, except that it usually ends with the down-the-nose speak to whomever your were arguing with.

Great work again, trying to make yourself out to be the smartest chap on the I-net.

-Al

Paul said...

Is there no end to blubber blather?

Allina said...

Lovely information ...Looking to improve your SEXUAL LIFE! Try Vigrx Plus the number one natural herbal erectile enhancement supplement for men.

Todd said...

Kurt,

Let’s take your definition of food reward:

“FR is the likelihood, given consumption of a food, that you will show behavioral motivation to eat more, whether you actually are allowed to or have the money to or if there is enough of it, etc..”

So FR is a “drive” or perhaps “drive enhancer” that would result in (excess) consumption of food so long as there are no additional impediments (lack of availability, inconvenience, cost, perhaps other social or psychological impediments). In that sense it is perhaps like other drives or urges like sex.

Drives don’t necessarily result in action. But they increase the probability of action, so long as they are not blocked by other factors or lack of enabling conditions.

I think that is what you are saying, but I don’t want to presume. So please let me know if I understand you correctly or not.

In that case, it is fair to ask: what is the nature of the “food reward” drive? Unless we give it a specific nature and mechanism, it becomes an empty explanatory factor that can be cited to explain anything. In the history of science there are some funny examples of this. For example, sleep was explained as occurring due to the “soporific factor”, which remained undefined. The ancient Greeks believed fire was caused by the release a supposed substance called “phlogiston”, which was never found. Ultimately, of course, experimental science replaced this with the correct theory of oxidative combustion. Without the evidence of genetics and paleobiology, the Darwinian theory of “survival of the fittest” could be accused of similar circularity or emptiness.

So I look forward to Stephan and you elucidating the actually mechanistic underpinnings of food reward. Stephan has hinted that his next post will do just that. (So far, I've seen only some very brief references to dopamine and opioids). Otherwise, the term “food reward” just an unempirical catch-all phrase that can be used in a manner to explain anything in a circular manner. The devil is in the details.

Todd

John said...

Steve_Roberts,

I experience the same after eating any high carb meal: I soon crave another. I know that for both sweet (bananas) and bland (potatoes), I have a tendency to gorge. Whether that is due to food reward, I don't know. My weight doesn't fluctuate with that behavior, but obviously it does in others, so... Emily Deans has written about anxiety in rats after eating certain things, which is pretty interesting.

Pete said...

"But we know that is a nonsensical criticism."

"Get it yet?"

"It is really not that hard to understand, unless you have some motivation for not understanding it or not wanting to believe it."

Once again, Kurt lives up to his name.

dim sum massacre said...

Stephan,

In addition to satiation (which I discussed in my previous comment), I'd like to suggest some other factors that may contribute to food reward:

complexity of flavor (foods with simple but strong flavors have higher food reward)

consistency of flavor (foods manufactured to always taste exactly the same have higher FR)

speed of eating (foods eaten quickly have higher FR)

Mike said...

If (for example) you believe excess fructose consumption causes weight gain by damaging the liver which negatively impacts insulin sensitivity, leading to hyperinsulinemia, leading to increased fat storage, leading to hunger and more eating, would that also be an example of food reward in that consumption of fructose eventually leads to motivation to eat more? Or would this and food reward be considered 2 separate mechanisms possibly working in tandem? If 2 separate mechanisms, is there a way to determine what proportion of the increased motivation to eat is driven by each? I.e., is food reward 5% or 50% or 95%? The same question could be asked for any diet.

Hopefully this thread hasn't degenerated beyond the point of useful discussion, though I would agree with Kurt (at least for myself) that the reason I am having a hard time understanding food reward is likely driven by a motivation to not want to understand it. If food reward is correct and is responsible in large part for my weight loss on the archevore diet, then it unfortunately moves me from being very happy and satisfied that my diet was full of wholesome/delicious/satisfying foods and that I wasn't really missing out on much, while still maintaining a healthy weight to having to admit that the only reason I'm maintaining my weight is because I'm depriving myself and I'd have to continue depriving myself for the rest of my life to maintain the weight. Clearly just a mindset shift, but one that could make this kind of lifestyle/diet much harder to maintain. I suspect this may be why a lot of us are resistant to food reward hypothesis.

dim sum massacre said...

Mike,
Yes, excess fructose is likely bad for you, and it's entirely possible that the metabolic damage it caused compounded whatever problems you had controlling your appetite.

But what caused the excess fructose consumption in the first place? In general, what is it that causes us to overconsume food? Why doesn't our appetite regulation system naturally stop us from doing this? That's where the food reward theory comes in.

Also, the term 'reward', when used in this context, really has a different meaning than the way we use the word in normal conversation. Eating a low food reward diet primarily means eating a diet that doesn't overstimulate the dompamine mediated reward pathways in the brain. And as Stephan notes, these reward pathways are actually separate from the brain pathways that determine whether an experience is pleasurable or not.

So it really doesn't need to be the case that a low food reward diet is less delicious or satisfying. In fact, I think that for most people a low food reward diet can ultimately be both more delicious and more satisfying than the SAD.

Anonymous said...

@Berto

Just skip my posts and my blog as well if they annoy you so much.

Aren't you the one who blogs while drunk and insults people out of the blue or was that some other crackpot?

Maybe that was a different Berto....

As far as narcissism goes, what do you call it when you think I am addressing you but I am not, yet you feel you have the right to insult me?

Did I address you or insult you? No, I did not, "SIr"

You are right I should not have expressed frustration in my response to Todd the way I did.

Todd, if you are reading this, I apologize for my tone.

But it frustrates me to see posters imply that Stephan and the researchers he references (and presumably me and Chris Kresser or anyone else who claims to understand it) are basically deluded or stupid for investigating something that is apparently just a vacuous rhetorical tautology.

I realize Todd is not saying that specifically, of course.

But the resistance to this idea mystifies me, and I think most of the objectors really are somehow offended by what they think it implies more than just being appropriately skeptical.

And regarding the vehicle performance analogy, where one variable can be a component of another but they are not the same, well if it makes not sense to you I can't help you any further.

Anonymous said...

@Dim Sum

"the term 'reward', when used in this context, really has a different meaning than the way we use the word in normal conversation. Eating a low food reward diet primarily means eating a diet that doesn't overstimulate the dompamine mediated reward pathways in the brain. And as Stephan notes, these reward pathways are actually separate from the brain pathways that determine whether an experience is pleasurable or not.

So it really doesn't need to be the case that a low food reward diet is less delicious or satisfying. In fact, I think that for most people a low food reward diet can ultimately be both more delicious and more satisfying than the SAD."

At least Dim Sum gets it.

Or is deluded by a meaningless tautology like some of the rest of us .

psychic24 said...

@Kurt Harris

" They could have the same nutritive value and macro ratios and one may only differ from the other on how much is stimulates subesquent intake."

I have one problem with that. Have you ever seen those late night or early morning news bulletins on CNN or whatever news channel, stating "Chocolate can help you lose weight" or "Potatoes are actually good for your figure"..etc. The point is, and what irritates me most, is that nutrition, the scientific study of it, is almost always discovering new proteins or substances (remember sprouted GABA rice and how it promised to raise growth hormone?) that can always shed light and discredit past mechanisms. The crux of what i'm saying is this: white potatoes and yams have similar macro ratios, they have similar nutrient composition, and yet white potatoes are ranked higher up on the satiety index-- you can say that is a perfect example of food reward or you can just wait for another new substance to be discovered and discredit all the work you put into food reward.

I'm also interested in presenting a personal example, even though i know they don't carry much weight around these scientific discussions. From all the talk of food reward my understanding has been that if you eat a food and it provides some kind of "positive" effect, you are more likely to repeat the action of eating said food. So how do you explain this: Whenever I drink a cup of coffee in the morning I don't feel the need to eat until lunchtime, as well as the rest of the day I have no over the top urges for foods/ I don't overeat. I compare this to days when I don't have coffee and subsequently have food cravings and tend to overeat. It's simple enough to say coffee is similar to a drug, but it's not super convincing. Caffeine is excreted rather quickly(6 hour half life? most people are up 15 hours daily, at least) and whatever drug-like effects it expresses can probably be mimicked by food ingestion (like increasing dopamine, stimulating insulin,etc).Just like there are tons of normal people (both normal weight and overweight) who can drink 1 cup a day, and not suffer any side effects when they go a day or more without the stuff--it doesn't explain what starts the progressive increase (both in regards to coffee and/or food intake).

A better theory seems to be regarding inefficient energy production. Hunger is easy to control, most people (including myself) eat for energy--a threshold much easier to overshoot when nothing seems to do the trick...
*i would be much more convinced if you said that people overeat junk foods because their lack of nutrients increases the degree to which their energy is lacking--not instead by stating food reward and palatability, which seems to have an incredible degree of individual variability (making it easy to discredit).

Anonymous said...

@Todd

Firstly, I apologize for the frustrated tone in my last response.

I think you are overstating things in a way that is unfair to some solid researchers, but I shouldn't take it personally.

"I think that is what you are saying, but I don’t want to presume. So please let me know if I understand you correctly or not."

I think you have in your recap.

"In that case, it is fair to ask: what is the nature of the “food reward” drive?"

I think Stephan is addressing that and intends to more in upcoming posts.

"Unless we give it a specific nature and mechanism, it becomes an empty explanatory factor that can be cited to explain anything."

I sympathize with the desire to know the mechanisms and pathway right away, but do not agree that until we know them in detail that it is an "empty explanatory factor"

Apropos of your history of science examples, what about Mendelian genetics? The rules of mendelian inheritance were worked out and were usefully predictive for many many decades before anyone had a clue what a chromosome was and about 75 years or more (IIRC) before DNA was discovered.

Was mendelian genetics an "empty explanatory factor" until we knew what the biology was at work?

"Without the evidence of genetics and paleobiology, the Darwinian theory of “survival of the fittest” could be accused of similar circularity or emptiness."

So I am guessing that you would have actually made this accusation had you read darwin's OTOS when it was published?

There was slim evidence beyond empirical observation and theorizing at the time. Certainly no definite mechanism to the standard you are demanding.

"So I look forward to Stephan and you elucidating the actually mechanistic underpinnings of food reward."

Don't look to me. I make no claim to being a FR researcher. I am here to see what Stephan says further about it. And in the meantime the idea makes sense enough to not be summarily dismissed.

Like natural selection as an idea based on empirical observation, the idea makes sense to me.

And I deny that it is circular or a tautology or meaningless without the full mechanisms know the same way I would grant that Darwin's theory deserved a hearing and not immediate dismissal because the mechanism was decades away from discovery.

"the term “food reward” just an unempirical catch-all phrase that can be used in a manner to explain anything in a circular manner"

That seems a very strong statement given what has been presented so far.

Anonymous said...

@Mike

"If food reward is correct and is responsible in large part for my weight loss on the archevore diet, then it unfortunately moves me from being very happy and satisfied that my diet was full of wholesome/delicious/satisfying foods and that I wasn't really missing out on much, while still maintaining a healthy weight to having to admit that the only reason I'm maintaining my weight is because I'm depriving myself"

But you are not depriving yourself of anything but excess calories and putative toxins, as long as you grant that FR is not purely or even mainly hedonic (palatability)

I don't feel deprived by my diet in the slightest, but I have no doubt there is something like FR effect going on both when it works and especially when it doesn't!



"and I'd have to continue depriving myself for the rest of my life to maintain the weight. Clearly just a mindset shift, but one that could make this kind of lifestyle/diet much harder to maintain. I suspect this may be why a lot of us are resistant to food reward hypothesis."

I think you are right that this is a reason for resistance. I think generally there may be too much belief in food quality relative to the simple parameter of overconsumption..

Being obese on paleo brownies or lemon bars may indeed be worse than drinking coca cola and eating snickers and just not eating as much. If true, this is very hard to swallow for the pale crowd

Eating low carb junk food or even low carb whole foods that are "delicious treats" and staying obese may be inevitable for some people unless they act to limit FR in more hard core ways. This idea is probably very threatening, if not intolerable, to some of the LC crowd, who can't stand the idea that total calories matter and if LC does not work, they will need to suffer the way smokers or those with problem drinking suffer -they must stop medicating their brains with food and eat more like the thin...

I initially found the idea threatening as well, as it calls into question the NAD hypothesis in its original form, or at least requires consideration of the idea that FR may be a cultural NAD.

But narcissist that I am ;), I am open to modifying my ideas to any degree necessary if that makes them work better and increases their explanatory power.

Coach said...

Dr. Harris,

"You are right I should not have expressed frustration in my response to Todd the way I did."

That's all this, and my others like it were about.

I accept your apology. Let's drive forward.

-Al

Terry said...

There's a rather well known diet that has been around for many year. I believe it started at Duke U or UNC, in North Carolina - called "The Rice Diet". From what I recall, it is almost completely comprised of carbohydrate with little fat or protein. The Rice Diet program claims to have helped people lose and maintain substantial amounts of weight. I am not sure if Stephan or any of the other researchers who comment here are aware of it - or whether they have ever reached out to the doctors and professional who run the program - but would this be another example of a program that might owe its success to bland foods and reduced calories, rather than manipulation of macronutrients?

Anonymous said...

@Berto

Interestingly, it turns out that narcissists are quite aware that they are perceived as narcissistic by others, so it does little good to point it out to them : )

Yes, let's move on.

Terry said...

Since dietary fat is one of the very best carriers of flavor, it stands to reason that many find low carb dieting pans more palateable and easier to adhere to.
Not to mention that also, many find them to be more satiating than high carb, low fat diets. Nevertheless, what I have picked up from this conversation is that this potentially could also lead to the undoing of these plans :(

sandra said...

I for one have no resistance to FR being correct, honest: I have been happy (no feelings of deprivation) on numerous diets - vegan, vegetarian, SAD, Paleo high fat, Paleo low fat... If anything switching to a "low reward" diet (at least the way I am understanding the explanation) would be nice as I'd save money on the food bill (sweet potatoes are delicious w or w/o butter, but that Kerry gold can add up!) Furthermore, though I've had no weight issues on any diet... I have had health problems. I cannot get past the notion that since DOC travel with obesity, whatever causes obesity may have caused my health problems even though I am not obese. Basically, I'll eat anything that will keep me healthy no matter what it tastes like!

I’m certainly grateful for all the work that Sefan does and for and his patience with those of us who are having trouble with the concept, but I also hugely appreciate Todd and others who I think are for the most part politely challenging Stefan... I would imagine that any honest scientist would welcome this challenge as it would give him the opportunity to demonstrate the strength of his theory!

For me it may help if someone could elaborate on the following:

“Eating a low food reward diet primarily means eating a diet that doesn't over stimulate the dopamine mediated reward pathways in the brain. And as Stephan notes, these reward pathways are actually separate from the brain pathways that determine whether an experience is pleasurable or not.”

What are some examples of “no-no” foods that over stimulate the dopamine pathway yet bypass the “pleasurable or not” pathway?

Also, how would we isolate FR effect from other effects that some might claim that food has so we can determine which has caused someone to eat more later (ei: if the food is a cookie is it FR messing with dopamine pathwys or it it the glucose spike).

dim sum massacre said...

Sandra,
To me the "no no" foods would be things that stimulate both the reward pathway and the pleasure pathway.

I think the best way to determine what these foods are is simply to pay close attention to how different foods affect your appetite, future hunger, and cravings. After you have taken a few bites of something, do you find yourself starting to feel satisfied, is the urge to eat more increased? Are there certain foods that you can't help overeating? How long does it take before you start to feel hungry again?

After doing this for a couple of months, I've found for example that, for me, a baked potato with butter is fine in terms of food reward. However a twice baked potato with sour cream and bacon bits and melted cheddar cheese isn't. But that's just me. Someone else who has a lower threshold for overstimulation might be better off eating their baked potatoes without anything.

With regard to your point about glucose spikes, my suspicion is that glucose valatility can actually serve to enhance food reward; ie, when our blood sugar drops quickly, our brain senses this and causes us to seek out food that will get our blood sugar back up. In the modern environment, these will tend to be high reward foods, which then overstimulate the dopamine pathway, leading to further appetite dysregulation, increased cravings for unhealthy food, more glucose spikes, etc.

psychic24 said...

i'm assuming my comment was a complete miss

Sara said...
This comment has been removed by the author.
Sara said...

@jpblogtest Hi! In fact what you say completely fits in with the reward theory because as you said 'carbs made you hungrier'. That is, you wanted more of them because they were rewarding to your brain. Remember, Stephan is not talking about reward in terms of your emotional reaction to foods, he is talking about brain chemistry. So, although I too, love to eat say.. a fried egg, and would say that I find it rewarding, actually it is not triggering the chemical cascade that represents 'reward' in my brain and I know this because when I eat a fried egg, I don't develop a craving for them. From what I've read, the test of the reward hypothesis is whether that food causes you to seek more of it. If it doesn't, then it is not hyper-rewarding, even if you like that food a lot.

Feralandroid said...

I may be missing something, but it seems to me that FR is not opposed to what Taubes was espousing in GCBC. From this article and the comments, it appears that it still comes down to certain foods that are consumed, instead of the quantity of your overall calories (CI = CO).

I do find the FR theory less informative then the Insulin theory or other theories based on the hormonal response of the body to certain nutrients.

Todd said...

@Kurt

“Firstly, I apologize for the frustrated tone in my last response.

I think you are overstating things in a way that is unfair to some solid researchers, but I shouldn't take it personally.
”

No worry, Kurt, I took no offense. I actually find your direct and passionate style refreshing! I must admit I’m a fan of Archevore, from which I’ve learned much. I’m similarly indebted to Stephan for his insightful blogging.

If my skepticism seems to be stubborn or pig-headed, I assure you it is sincere, not motivated by a desire to reject the food reward theory. In fact, I’m quite sympathetic to certain aspects of the food reward theory, particularly the interest in clarifying the connections between neurobiology, appetite, flavor, and obesity. I believe that obesity is associated with dysregulation of hypothalamic processes. I’ve explored these connections for some time on my blog “Getting Stronger” (http://gettingstronger.org/). What I’m not convinced of (yet) is the idea that the obesity problem starts with certain types of food, such as those on the “cafeteria diet”, which Stephan considers “high reward” foods. I think we first need to understand how it is that some people are much more vulnerable than others to the siren song of high reward foods. I suspect it starts with a combination of genetic susceptibility and metabolic “damage” resulting from a poor diet and inadequate exercise, leading to leptin resistance and hypothalamic insulin resistance. Once damaged, the brain then becomes susceptible to high reward foods, reinforcing obesity. But most likely, obesity doesn’t start with the food reward "drive". The "drive" comes later.

So I hold something like the “mirror image” of Stephan’s reward hypothesis. I accept most of the data establishing the correlation between neurobiology and obesity, but my main argument is with the causal sequence. At this point, I consider my view to be a well-informed hunch based on what I’ve read. But I assure you that I’m more than willing to revise my position and accept the food reward hypothesis if Stephan, you, or anyone else can provide credible empirical evidence. My mind is open.

Darwin’s theory of evolution was always more than a general schema based on “survival of the fittest”. Well before Watson and Crick, or even Mendel, Darwin had marshalled volumes of empirical evidence and detailed arguments to support his theory, including biogeographical findings from the Galapagos islands, and breeding experiments. With every decade, evolutionary theory has become increasingly well validated by confirmation from fasifiable experiments.

What I’m looking for from proponents of the food reward hypothesis is enough empirical evidence to at least support the idea that food reward is primarily an instigating cause, rather than an effect, a correlate, or a merely sustaining force, for obesity.

Regards,

Todd

Todd said...

@Kurt

“Firstly, I apologize for the frustrated tone in my last response.

I think you are overstating things in a way that is unfair to some solid researchers, but I shouldn't take it personally.


No worry, Kurt, I took no offense. I actually find your direct and passionate style refreshing! I must admit I’m a fan of Archevore, from which I’ve learned much. I’m similarly indebted to Stephan for his insightful blogging.

If my skepticism seems to be stubborn or pig-headed, I assure you it is sincere, not motivated by a desire to reject the food reward theory. In fact, I’m quite sympathetic to certain aspects of the food reward theory, particularly the interest in clarifying the connections between neurobiology, appetite, flavor, and obesity. I believe that obesity is associated with dysregulation of hypothalamic processes. I’ve explored these connections for some time on my blog “Getting Stronger” (http://gettingstronger.org/). What I’m not convinced of (yet) is the idea that the obesity problem starts with certain types of food, such as those on the “cafeteria diet”, which Stephan considers “high reward” foods. I think we first need to understand how it is that some people are much more vulnerable than others to the siren song of high reward foods. I suspect it starts with a combination of genetic susceptibility and metabolic “damage” resulting from a poor diet and inadequate exercise, leading to leptin resistance and hypothalamic insulin resistance. Once damaged, the brain then becomes susceptible to high reward foods, reinforcing obesity. But most likely, obesity doesn’t start with the food reward "drive". The "drive" comes later.

So I hold something like the “mirror image” of Stephan’s reward hypothesis. I accept most of the data establishing the correlation between neurobiology and obesity, but my main argument is with the causal sequence. At this point, I consider my view to be a well-informed hunch based on what I’ve read. But I assure you that I’m more than willing to revise my position and accept the food reward hypothesis if Stephan, you, or anyone else can provide credible empirical evidence. My mind is open.

Darwin’s theory of evolution was always more than a general schema based on “survival of the fittest”. Well before Watson and Crick, or even Mendel, Darwin had marshalled volumes of empirical evidence and detailed arguments to support his theory, including biogeographical findings from the Galapagos islands, and breeding experiments. With every decade, evolutionary theory has become increasingly well validated by confirmation from fasifiable experiments.

What I’m looking for from proponents of the food reward hypothesis is enough empirical evidence to at least support the idea that food reward is primarily an instigating cause, rather than an effect, a correlate, or a merely sustaining force, for obesity.

Regards,

Todd

P2ZR said...

Todd,

So after this entire exchange you've had with Kurt (which I admit I haven't examined in full detail), are you saying that your "mirror image" of the FRH--that after becoming damaged, the brain and/or body is more 'susceptible'--is pretty much what Paul Jaminet is saying?: (http://perfecthealthdiet.com/?p=3704; http://perfecthealthdiet.com/?p=4268)

I'm a fan of much of Paul's work, and find these ideas interesting, but here I'm more inclined towards an Occam's razor line of reasoning.... While there's obviously a neurobiological basis for what psychologists term "addictive personality", I'm not sure those people's dopamine pathways were necessarily weirded up in the first place by malnutrition and/or toxins (maybe genetics). Methinks they will just be more 'susceptible' to highly rewarding food, even if they were brought up on fairly wholesome fare....

Asim said...

http://www.sciencedaily.com/releases/2007/10/071025091036.htm

""This research corroborates brain-imaging studies conducted at Brookhaven that found decreased levels of dopamine D2 receptors in obese people compared with normal-weight people," said Brookhaven neuroscientist Panayotis (Peter) Thanos, lead author of the current study, which will be published in the journal Synapse.

It's not clear whether reduced receptor levels are a cause or consequence of obesity: Overeating may chronically reduce receptor levels, which, over the long term, could eventually contribute to obesity. But having genetically low receptor levels may also lead to obesity by predisposing the individual to overeating in an attempt to stimulate a "blunted" reward system. Either way, revving up receptor levels by restricting food intake could enhance the impact of this common strategy for combating obesity."

Asim said...
This comment has been removed by the author.
Asim said...

What this seems to be implying is the following:

You can be obese if your food-reward system isn't functioning properly, i.e. your not getting enough reward value out of ANY food you are eating, thus you may be over-eating in an attempt to stimulate a reward...

TCO348 said...

I think what is at issue is the following :

Do the sensory stimuli of eating, in and of itself, either have an effect on fat-mass setpoint or have the capacity to override one's appetite and satiation, or both, or does it not have either of these effects?

I believe that the sensory stimuli do have these effects. I believe that the neural signals coming from the taste and smell organs of the mouth and nose affect the inclination to eat independently of any metabolic effects.

I believe that we evolved to find certain tastes and smells pleasurable so as to make us more inclined to eat and drink those things that give us the experience of those tastes and smells because the consumption of these things improved our chances for survival.

I believe that there is a strong inclination to experience pleasure and when the opportunity to experience pleasure is available to us at little cost we will take advantage of it.

I don't believe that this pleasure is only derived under certain metabolic conditions. I understand that it is well know that specific appetites for various nutrients can vary based on our need for those specific nutrients and the amount of pleasure derived from consuming them varies accordingly. This does not imply that our capacity to derive pleasure from foods is entirely dependent on a metabolic need for those foods. I believe that we can continue to derive pleasure from the foods that strongly stimulate our strongest pleasurable taste receptors even when there is no deficiency, especially if those foods are stimulating several such pathways at one time.

I don't believe that the existence of any metabolic damage is necessary for there to be this inclination to eat highly pleasure stimulating foods. I believe this is true even when a person is completely sated and satiated.

It seems that Todd doesn't agree with this. He thinks that sensory input from eating has no independent effect on inclination to eat beyond caloric needs or to raise the fat-mass set-point.

Asim said...

http://addiction-dirkh.blogspot.com/2008/04/food-addiction-and-dopamine.html

"In the study, Ivan de Araujo and colleagues at Duke University and the Universidade do Porto in Portugal demonstrated that lab mice lacking the ability to taste sweet foods still preferred sugary water to regular water. The genetically altered mice, lacking functional taste receptor cells for bitter and sweet, consistently chose to consume sugar water--even though they could not sense the sugar. (The lab animals were also prevented from smelling or sensing textural differences in the offerings.)

"Our findings suggest that calorie-rich nutrients can directly influence brain reward circuits that control food intake independently of palatability or functional taste transduction," the researchers wrote.

The findings offer new clues to obesity, and also bolster the contention that simple carbohydrate foods--because of their effect on reward pathways in the brain--can be addictive for certain people. As Tamas Horvath of Yale University's School of Medicine told Science News (sub. required): "This is a very exciting new element in how you get addicted to food. It doesn't even matter how it tastes."

Asim said...

Another interesting comment from the same article:

In the same article, written by Amy Maxmen, study author de Araujo said: "The animal's reward processing systems were sensitive to changes in metabolism, not just flavor. This is a new system."

The "sweet-blind" animals did not go for the low-cal alternative, when they were offered water mixed with sucralose, otherwise known as Splenda. Low-cal sweeteners did not result in a similar dopamine boost along the reward pathways of the brain.

TCO348 said...

Hunger and Palatability :

Let me propose a particular way of defining hunger just for the sake of this particular argument. I don't intend for this definition of hunger to supplant all others but it is useful for this discussion.

So, just for this discussion, hunger is the point at which someone thinks "I'd really enjoy eating something right now".

It is well known that the hungrier you are the better your food tastes. When, for whatever reason, you are forced to go a long time without eating the anticipation of the meal is intense and it always tastes great. 'Hunger is the best sauce' as the saying goes.

Now consider a hunter-gatherer. They have never experienced professionally prepared food that hits all the taste receptors at once. Salt, fat, tart, umami, sweet, great aroma, all in the same bite. Think pizza. In their memory banks of food consumption are the foods they typically eat. Their unconscious mind is going through the available options and asking ' would that be good right now'? If the available option is some leafy greens or some nuts they will need to be considerably hungrier on a physiological level for the answer to be yes. But they won't feel any hunger pangs until they get to that point. They know unconsciously what is available and they are not at the point where those foods would be pleasurable yet so they have no inclination to eat.

Now contrast that with the situation in the US where just about anything you want is at arm's length. It won't take as much of physiological nutrient deficit to get thoughts of what foods would be good to start popping into your mind.

If you have a policy not to eat super-palatable food then you will not consider these available and you will not experience cravings for them. But it takes some doing to get to that point. (I'm not there right now but I have been in the past)

Also, keep in mind that hunter-gatherers do seek out food. They do derive pleasure from it. A person must or they will die. The idea is not to completely remove any and all pleasure from eating. Its to find the right amount of pleasure. The amount that allows a healthy weight.

PS - Mike - I think you're exactly right about why people are resistant to the food reward idea.

Todd said...

@ Sarah Barracuda

Yes, I agree with much of those posts of Jaminet's, though I may not be quite as strict about a pure paleo diet as he is. I think this statement of his nails it on the head:

"...it seems likely to me that food reward becomes a dominant factor in obesity only after some form of metabolic damage from malnutrition, toxins, or infections begins to affect the brain’s regulatory systems. In a healthy person a highly palatable diet might have little effect on weight for quite some time...Distinguishing my view from Stephan’s is difficult because the obesity-inducing diets used in animal studies are generally both toxic and malnourishing and highly palatable. The “cafeteria diet” of Cheetos and such – rich in wheat, sugar, and vegetable oil – is an example."

@ Tayor

"I don't believe that the existence of any metabolic damage is necessary for there to be this inclination to eat highly pleasure stimulating foods. I believe this is true even when a person is completely sated and satiated."

There is plenty of evidence that being completely sated or "full" can dramatically reduce the drive to eat. Stephan wrote about this pheonomenon of allesthesia here: http://wholehealthsource.blogspot.com/2011/08/food-palatability-and-body-fatness.html

If you think that foods which are highly palatable to most Westerners have a universal power to ignite overeating and obesity, what about the following two counterexamples:

1. Seth Roberts writes Gaku Homma, a Japanese cookbook author whose initial impression of Coke was that it tasted “like medicine” and was repulsed by it. So he had zero inclination to consume a beverage that Westerners find rewarding. Similarly, Westerners are often repulsed by Asian fermented foods, or delicacies like dog or snake. Certain cultures find insects and grubs to be delectable, but most of us would probably pass, even knowing that such foods represent a nutritious source of calories. "Food reward" is culturally relative.

2. I used to love soda and highly spiced potato chips. After several years of low carb/paleo dieting, intermittent fasting and intense exercise, I now find such foods repulsive. I spit out coffee that is sweetened. "Food reward" is individually variable, and changes over time.

Todd

TCO348 said...
This comment has been removed by the author.
TCO348 said...
This comment has been removed by the author.
TCO348 said...

There is plenty of evidence that being completely sated or "full" can dramatically reduce the drive to eat.

I don't doubt this. But the better something tastes the longer you will eat it. Some things taste so good they are pleasurable to the point of physical discomfort from being overstuffed.

As for changing tastes or culturally different tastes - that doesn't imply that metabolic dysregulation is necessary for a food to drive overeating. I didn't like cola as a child. I didn't like hot dogs. I didn't like Indian food when I first tried it. Food preference is complicated but I don't really see how that implies that hyperpalatability can't drive overeating in the absence of metabolic dysregulation. Perhaps we have an built-in aversion to unfamiliar tastes and we come to like them more after we try them a few times. Perhaps we become sensitized to sugar when we don't eat much of it and sweet things taste too sweet then. None of these phenomena imply that metabolic dysregulation must occur first before hyper-palatable food can drive overeating. Why do you think that metabolic dysregulation must precede overeating on hyperpalatable food?

The fact that not all people are susceptible to it also doesn't necessarily imply that metabolic dysregulation accounts for the difference. There are many possibilities. I'd bet on genetically based differences in opiate and dopamine receptors. I don't know this but I don't see why you assume that the difference must be due to metabolic damage caused by junk food.

sandra said...

@dsm:
Thanks for your comment.

I'm still wondering about the following though:

"To me the "no no" foods would be things that stimulate both the reward pathway and the pleasure pathway."

Stefan has also written that reward and palatability often travel together..

But if the two often travel together, and the most problematic foods do both (reward and pleasure), why can't we JUST talk about palatability. That is why I asked for an example of a food that would stimulate the dopamine pathway but not the pleasure pathway...if you cannot separate them in the real world then wouldn't it just boil down to people tending to overeat tasty foods more so than foods that are not so tasty (just as most people would rather be hugged than pinched)?

I do also have foods that I think fit your description of high reward: one for me is chocolate. I do get cravings for it if I eat too much or consume it too frequently...BUT, I stay in calorie balance even when "overeating" it. So this HR(and very palatible) food makes me eat more of it alone, not more of all foods. In fact, I guess I must be eating less of other foods to compensate as my weight remains stable. I cannot see how this kind of craving can explain the obesity epidemic.

Todd said...

@ Taylor

"The fact that not all people are susceptible to [food reward] also doesn't necessarily imply that metabolic dysregulation accounts for the difference. There are many possibilities. I'd bet on genetically based differences in opiate and dopamine receptors. I don't know this but I don't see why you assume that the difference must be due to metabolic damage caused by junk food."

I agree with you that metabolic dysfunction is only one possible avenue to high food reward susceptibility. In my earlier post I did mentioned genetic predisposition as being a key factor. I would add psychological conditioning as another important factor in susceptibility.

The common thread here is that vulnerabilty to obesity stems from a brain that is wired in a way that inhibits normal appetitive feedback and exhibits leptin resistance and hypothalamic insulin resistance. Genetics, diet and conditioning can each influence this wiring.

But many (myself included) are wired in a way that we either lost, or never had, this response to the same foods that drive others to overeat. Food reward seems to me not to be a drive that can get under your skin without first having a susceptible brain and metabolism.

TCO348 said...

Food reward seems to me not to be a drive that can get under your skin without first having a susceptible brain and metabolism.

I'd make the 'and' an 'or'. I don't see why the metabolism has to be any different than normal. If a person (me for example) is tempted to have a slice of pizza every time I pass a good pizzeria I don't think that says anything at all about my metabolism. I think it says I really loves me some pizza. I'm not sure why you think the metabolism must be involved.

Coach said...

Dr. Harris,

Please... but only if you call me "Al". "Berto" is part of my google account name.

Thanks Dr. Harris, for everything that you do.

-Al

dim sum massacre said...

Sandra,
My own feeling on this (Stephan might disagree) is that reward and palatability absolutely can and should be separated. Palatability is one of the components that goes into determining food reward, but because there are other factors that are equally important, palatability alone is not sufficient.

If we define palatability as tastiness, it just doesn’t seem credible to me to say that the French and the Japanese are less obese than Americans because their diets are less palatable. I’ve been to those places, let me tell you, their food is damn tasty! More likely, it’s that their diets hit the sweet spot of being both high palatability and low reward. Thus they tend not to overeat, despite being highly epicurean, food-obsessed cultures.

That’s what I was trying to get at by making the comparison between potato chips and fois gras in my first comment in this thread. Potato chips are high in both palatability and reward, but fois gras is an example of a food that has high palatability without high reward.

Chocolate is another interesting example. I find that for me, the reward value of chocolate correlates strongly to the amount of sugar added. I can eat one or two squares of a 80% cacao chocolate bar, enjoy it tremendously, and put away the rest of the bar without having any urge to eat more. However if I eat something like a Hershey’s Kiss, I actually find the taste weird and overly sweet and don’t enjoy it nearly as much. And yet, a couple of minutes later, I have the strange urge to go back for more. For me at least, adding sugar to the chocolate simultaneously lowers palatability and increases reward.

Of course when I was a kid I loved Hershey’s Kisses, and probably wouldn’t have cared for the taste of 80% cacao. Which points out the fact that palatability is partially a learned response, and can change over time and in response to what we eat.

It seems like processed foods with lots of added sugar and fat actually teach our brains to associate high palatability and high reward.

Ed said...

Didn't read all 114 comments (wow), and dunno if anyone will read this (shrug) but one line of thinking that's not jumping out to me is hysteresis. The path to obesity may not be the same path in reverse to leanness.

Based on society-level data, it strikes me that it is very plausible that the food-reward hypothesis explains the massive increase in obesity in the late 20th century.

However, the obese condition it creates, may require something beyond low-reward food to reverse.

One line of thinking that is popping up is impaired mitochondrial function. It strikes me as plausible that a low carb diet might be more supportive of health if your excess calories have caused mitochondrial dysfunction.

In any case, like many biological states, there is more than one path to obesity, and more than one path in return. I understand Stephan is trying to explain the massive societal shift that we are seeing. When we have a massive societal shift back to leanness, perhaps we can better talk about the dominant factors in recovering from obesity.

J. said...

This is something I keep fearing&struggling with,which leads me to become even more eat-disordered.

However there are so many factors that come into play according to all the varies studies: Food reward,messed-up brain&hormonal chemistry,bad gut flora,other parasites etc. in the body,vitamin&mineral deficiencies,MSG/Glutamate foods,emotional stuff,sleep deprivation etc.