Friday, August 9, 2013

Food Reward Friday

This week's lucky "winner"... cola!

Thirsty yet?  Visual cues such as these are used to drive food/beverage seeking and consumption behavior, which are used to drive profits.  How does this work?  Once you've consumed a rewarding beverage enough times, particularly as a malleable child, your brain comes to associate everything about that beverage with the primary reward you obtained from it (calories, sugar, and caffeine).  This is simply Pavlovian/classical conditioning*.  Everything associated with that beverage becomes a cue that triggers motivation to obtain it (craving), including the sight of it, the smell of it, the sound of a can popping, and even the physical and social environment it was consumed in-- just like Pavlov's dogs learned to drool at the sound of a bell that was repeatedly paired with food.

This is a big part of why food advertising works.  Showing you food on TV or on a billboard is simply a cue that triggers your previously formed reward association, motivating you to obtain the food/beverage when it may not have crossed your mind otherwise.

Cola is an interesting beverage with a long history in the US.  It became popular after the development of Coca-Cola by the American pharmacist John Pemberton in 1886.  If you asked 20 people on the street what is used to flavor and color cola, I suspect most of them would have no idea, or guess some sort of industrial chemical concoction.  In fact, the flavor comes from a blend of common spices we're all familiar with, including citrus rind, cinnamon, and vanilla, and the color comes from caramel.

Of course, it also comes with a hefty dose of sugar, or more precisely, high-fructose corn syrup.  One 12 oz can clocks in at 136 calories, about 50% more than a medium apple.  A 20 oz beverage, also commonly consumed, weighs in at 226 calories.  Soda is nutritionally bankrupt, tends to be consumed carelessly on top of a normal diet, doesn't contain the substances (fiber, protein) that would normally limit intake and provide health benefits, promotes tooth decay, and is rewarding enough that we drink it in the absence of calorie need (hunger).  Sugar-sweetened beverages (as opposed to most other sources of sugar) are linked to the development of obesity in observational studies and controlled trials, although the evidence hasn't always been consistent.

Why do we drink soda?  Because it's rewarding, almost by definition (reward = motivation and behavioral reinforcement).  If it weren't rewarding, we'd just drink water instead.  Why is it rewarding?  No mystery here.  As with all hyper-palatable/rewarding foods, it's somewhat of a perfect storm.  It contains a concentration of sugar that's right at the human "bliss point" that is consistently exploited by food manufacturers-- not too low and not too high.  It contains caffeine, a habit-forming drug (and it used to contain cocaine).  It tastes good and has a pleasant fizz.  Last but not least, it's easy to ingest.

* The calories, sugar, and caffeine are the unconditioned stimuli; the appearance, aroma, atmosphere, etc. are the conditioned stimuli.  Once conditioned by repeated pairing with the unconditioned stimuli, the conditioned stimuli become cues that drive motivation for the beverage.


Anonymous said...


A 12 ounce can has 140 calories, not 97. A 20 ounce bottle (so common today) approaches 250 calories.

Phil Koop said...

If I may be tangential for a moment, I have reservations about your attempts last week to be "less extreme" and move away from foods that are "easy to laugh at".

It may be easy for the readers of this blog to laugh, but that is selection bias; such readers are an irrelevant minority. The reality is that millions of people are actually consuming these foods. It is practically impossible to be too extreme, as you can see from this review of the Indiana Food Fair.

Which brings me back to the subject at hand. The egregious foods of the Indiana State Fair are not only harmful in themselves, they also shift the Overton window for what "seems reasonable". For example I note that you have chosen 12oz as a representative serving size. Compared to servings of 20z and more that are now available, this does indeed seem moderate.

However, although must admit that I sometimes enjoy drinking cola, if I pay attention to how much my appetite actually "wants", it is almost always much less than 12oz. Those little 6oz cans are perfect.

raphi said...

Yes it made me thirsty, but in my defence my N=1 is confounded due to the inebriating heat in Nice (S. of Fr) at the moment :)

Do you think the 97 calories have a detrimental impact on metabolism because they're 'empty' or do you think the impact has more to do with what happens when your body deals with however many grams of sugar are in those 97cals?

Whether we try to explain the ill health effects of soda consumption more as a quantitative and energetic issue instead of a qualitative hormonal one, are you saying sugar is still not the major factor in the equation?
If not, then what's our next best guess?

Thanks Stephan!

Stephan Guyenet said...

Hi Mark,

Oops, good catch! I corrected the post.

Hi Phil,

Good points.

Hi Raphi,

I was too hasty in looking up the calorie value-- it's actually 136 per 12 oz can. I haven't seen any evidence that would lead me to believe that sugar itself has any special impacts on the metabolism that favor fat gain (vs. other refined calorie source), in the amounts that are consumed in a normal diet. As far as I can tell, the key problem with sugar and body fatness is it makes you eat more, and the palatability of certain sugar-containing foods may indirectly affect energy homeostasis.

Charlie Currie said...

Growing up poor and happy in the 1950s, our family would celebrate Friday night with a glass of Pepsi and a bowl of popcorn while we watched TV.

That was the only time we had soda/pop or soda pop of any kind - once a week, maybe, and it was probably 6 oz or less for my sister and I.

I've never been a big soda drinker, the carbonation always bloated me up and as a kid I could never burp like my buddies - consequently, I just suffered.

I haven't had a soda in probably five years - last Sunday on my wife's and my quest for the best burger in Long Beach, we were at a little bar near UCLB and they had artisanal draft root beer on tap. Because everyone was talking about how good it was I got myself a little sample. It was amazing - sparkly on the tongue and real smooth and creamy going down - no bite or bitterness - no esophageal burn. It was good enough for maybe a return trip for one of their bbq'd dogs and a glass to wash it down.


Clint said...

I've been thinking about how diet sodas seem to be associated with weight gain in observational studies. Now, I don't think diet soda by itself can cause weight gain. I think if you had a cohort in an observation study that ate a fairly bland diet and also drank diet soda, diet soda wouldn't be associated with weight gain in that cohort. But what happens when you pair diet soda with palatable food (i.e. fast food)? Does the caffeine and sweetness of diet soda increase the reward value of such meals? What if people ate caffeine pills every time they ate palatable meals? Would caffeine pills be correlated with long term weight gain? On the other hand, coffee and unsweetened do not seem to be associated with weight gain in observation studies even though they also contain caffeine. Is it the lack of sweetness and/or the fact that these beverages are often consumed in isolation (i.e. many people simply have coffee in the morning)?

detroit dog said...

I drink a Pepsi when I have a migraine coming on. I'm not sure why, but it seems to help.

When I was a kid, our Mom gave us a spoonful of coke syrup when we were nauseous. Usually, it helped me "toss my cookies" and I felt much better afterwards.


raphi said...

Doesn't the fact that sugar "makes you eat more" plainly suggest other foods make you eat less, thus demonstrating its distinct metabolic influence by simple comparison?

I'm curious about what you mean by palatability directly affecting energy homeostasis...Are you thinking in terms of palatability ---> affecting our choices --->and in turn leading us to ingest more calories (often with a poorer calories-to-nutrients ratio)?
Or do you mean our direct conscious experience of food (I guess our emotional response) can be sufficiently profound to affect a significant metabolic change (either negative or positive)?

SimonM said...

In my RSS reader I read the first line as This week's lucky "winner"... coal!

Thought you'd run out of the more obvious ones rather quickly.

You know what they say: a geologic eon on the lips, a lifetime on the hips.

Stephan Guyenet said...

Hi Raphi,

The fact that a food makes you eat more or less does not necessarily suggest a metabolic effect. The sensory characteristics of food can influence food intake independent of any post-ingestive effect.

Regarding the effect of reward/palatability on energy homeostasis, I've written about it at length on this blog as well as in the scientific literature. See my posts on food reward, or my JCEM review paper (link in sidebar).

Puddleg said...

Coca specifically contains coca alkaloids and other phytochemicals. The cocaine is removed (traces remain) but similar alkaloids can be retained legally. Actually the legal description of the de-cocainated extract is rather dated and unsatisfactory from a pharmaceutical perspective.

Anonymous said...

Alas, all is never what it seems. Replacing sugar sweetened beverages in teenage diets reduces weight in year one, but after two years weight is unchanged. My point in referencing this article isn't to suggest that Stephan isn't onto something, but I just think one needs to be careful before demonizing cola. The fattest people I know don't drink any sugar sweetened beverages.

A Randomized Trial of Sugar-Sweetened Beverages and Adolescent Body Weight by Cara B. Ebbeling, in the October 11, 2012 issue of the New England Journal of Medicine, found that “Among overweight and obese adolescents, the increase in BMI was smaller in the experimental group than in the control group after a 1-year intervention designed to reduce consumption of sugar-sweetened beverages, but not at the 2-year follow-up (the pre-specified primary outcome).” Cara B. Ebbeling, Ph.D., Henry A. Feldman, Ph.D., Virginia R. Chomitz, Ph.D., Tracy A. Antonelli, M.P.H., Steven L. Gortmaker, Ph.D., Stavroula K. Osganian, M.D., Sc.D., and David S. Ludwig, M.D., Ph.D. A Randomized Trial of Sugar-Sweetened Beverages and Adolescent Body Weight. N Engl J Med 2012; 367:1407-1416

Unknown said...

There's nothing inherently wrong with a cola or any other soda that I'm aware of.

Stephan I completely agree that there is no compelling evidence that sugar, or HFCS or any other carbohydrate has any meaningful affect on an increase in DNL versus any other macro nutrient or carbohydrate specifically. DNL is not the major contributor to TGA accumulation in adipose, it's dietary fat and positive energy balance.

It continually comes down to food reward and a psychological shift towards many peoples view of being fat coupled with abundance and being more sedentary. Correct me if I'm wrong Stephan but this is the center of your research and similar to your hypothesis.

aluchko said...


There's been some research suggesting that artificial sweeteners induce overeating. Basically the sweet flavour primes the body to injest calories, then when no calories show up the body gets hungry looking for them.

This could suggest the association of diet pop with weight gain.

That being said I'd be curious to see how the effect actually works, for instance zero calorie pop might be bad, but veggies with artificial sweeteners to increase the paletability is maybe good.

Anonymous said...

I'm a sucker for classical conditioning.

I use to stop at a 7 Eleven on my way to or from a squash court to pick up a big slurpee of diet cola. Every time I even drove past, I felt that craving. However, it is possible to "decouple" the stimulus. It does take time and effort. Now when I pass the 7 Event on my way to play squash I only very rarely think of getting a big slurpee and I never do. It seems to me that the trick in overcoming any craving is to stop the cycle of craving by not giving in to the "food reward". Giving in only reinforces the cycle. Total abstinence is the only way. Ajahn Chah and other Buddhists have noticed this. He used to call craving "wide open". There is absolutely no way to satiate it in the long run - other than to cease it.

I wish this would work with harder chemical addictions like heroin and alcohol but I think it only rarely does.

I am not one of those people who can enjoy something like this in moderation, like some of your previous commentators. One cola drink and I am sure I would fall off the wagon and be drinking it every day. Yes, I know I am talking about diet cola, but still.... isn't it better to have fewer cravings than more cravings?

Love and enjoy your blog, Stephan.


JBG said...

Re 12 oz vs 20 oz, in the 1940s the typical serving size was 6 oz. Pepsi led the way up:

Pepsi Cola hits the spot
Twelve full ounces, that's a lot
Twice as much for a nickle,too
Pepsi Cola is the drink for you!

Re the statement that soda "tastes good", soft drinks were not part of my childhood household. Nobody made a big deal about it, they just weren't around. So when at the age of twelve or so, friends urged me to try Coke, I had no prior experience. Blech!! I thought it tasted awful, and kidding from my friends wasn't enough to make me willing to drink it. My guess is that any person truly naive to the genre would react much as I did.

Anonymous said...

Dhackam -

I couldn't agree more. Abstinence is the key.

It's ironic that Nancy Reagan's "Just Say No" campaign, which even I laughed at as a teen, was perhaps the right prescription for not just drugs.

Mike said...

Great post as always. Thank you.

You blew past the "fizz" part, but that may be a big part of the reward and addiction.

Carbonation is registered as pain by the mouth, which triggers the release of rewarding endorphins.

Robert said...

I apologize for this off-topic comment, Dr. Guyenet, but is it possible that the genetic mutation driving uncontrolled appetite and obesity in Prader-Willi Syndrome (mutation known to affect hypothalmus) is also at work in other forms of obesity?

Here's a BBC video on PWS.

Unknown said...

Hi Stephan, this question isn't really related to food reward friday, but i'm interested in hearing your thoughts on the topic. Are you familiar with Miki Ben-dor? He is a "Ph.D candidate at the department of Archaeology of Tel Aviv University"

and his blog is

This is a quote from a blog entry..
"Here is my hypothesis: In order to maintain a dynamic optimum the body uses cues from the environment to gauge the probability that famine or abundance is imminent. Since meat is the preferred source of calories for humans, ingestion of plenty of meat is translated into a message of abundance and ingestion of larger than normal quantities of non-meat food is translated into a message of higher probability of famine.

In other words the body interprets ingestion of sub optimal foods as an indication of a shortage of optimal foods and begins to accumulate fat. Sensing plenty of optimal foods, the body reduces fat accumulation to a minimum in order to allow for optimum DEE and efficient chase or flight."


Anonymous said...

Mark, thank you for your kind comments.

I've come to believe there are two types of people: those who can consume only moderate amounts of chemically addictive substances over the long-term and those who cannot.

I clearly fall into the latter camp. The only way I could stop addiction to something like diet pop was to cease consuming it altogether. Just one beverage is my tipping point.

There is something amazing about the body's expectations of reward. On Saturday I had a long nap around noon and on Sunday my body expected me to do exactly the same, and was extremely tired when I did not (at exactly the same time of day). Today, Monday, a little less tired but no more willing to give in. By Tuesday, I will be even less tired, no doubt, and this will reinforce the urge not to give in. From a Buddhist perspective, life is all about releasing craving and realizing the cessation of craving (the central cause of suffering). If my body expects and demands a nap every day, it will suffer if I deny that, because it is craving it. But it will suffer a bit less each day as it no longer comes to expect the nap.

Habits can be broken. It takes time to break them, but it certainly can be done. The underlying knowledge and education is probably more important than the physical and mental willpower that is so touted in our society. Now I'm supremely off topic....

Stephan Guyenet said...

Hi Tom,

That hypothesis relies on untested assumptions that are unlikely to be correct. For example:

The body uses food type (independent of calorie intake) as a gauge of the probability of famine. I'm not aware of any evidence to support this.

"Meat is the preferred source of calories for humans". Not aware of any evidence to support this. I would like to see the definition of "preferred" here. Certainly humans enjoy meat and it's nutritious, but there is no evidence whatsoever that eating 60% of calories from meat is physiologically preferable to eating 20% of calories from meat.

There are far too many lean cultures eating diets very high in carbohydrate (60-80% of kcals) and low in meat for me to believe this.

Anonymous said...


I fall into the latter camp also. Cigarettes and sugar were my worst vices.

Very interesting regarding naps. I have found the same thing, but in an opposite fashion: If I work several days on 4 to 5 hours of sleep, I find my body adjusts and almost shuns sleep. Perhaps it gets addicted to adrenaline - I don't know.

The Russians have a saying: appetite begets appetite.

Perhaps they were onto something.

Unknown said...

Mark, the Russians may have it right imo.

I've said for years, (and experienced therefore..) that the more you feed the body the more it seems to expect. It's up to Stephan and others like him to figure out that mechanism but it certainly seems true for many.

The opposite is also true given a short amount of time! The body is very adaptive!

I've always joked about writing a diet book that is only a couple of pages long--the."Eat The Foods You Don't Like" book! If someone truly wants to lose weight simply order or prepare foods you don't care for!! (liver for me for example) and I promise you you will lose weight...for a while!

I predict given a few weeks those foods will become foods you enjoy and that effect will no longer exist. Then the foods you thought you couldn't live without will suddenly not be quite as palatable as you once thought.

This idea is not new but is telling! Ultimately it's very complicated and there is much more going on than most can conceive, while at the same time it's quite simple if you have the tenacity, or you don't have the hormonal imbalances, which not all fat people do.

In the end the basics must be covered first, CICO is a major part of the equation and the "Golden Egg" is going to be determining how to keep people from getting to the point that they become so dis-regulated that they end up at the point of no return. (CICO doesn't matter).

The problem in my mind is that this is so subjective that we may not get there. And pharmacological intervention doesn't seem to be the answer to me, with the exception of those that simply have the genetic disposition for being obese.

There have always been Friar Tucks and there always will be. Those folks are the ones we should help first in my opinion.

Puddleg said...

@ Mark Rand,

Hepatic DNL from carbohydrate produces 5% of serum TG normally BUT this rises to around 25% when there is NAFLD (fatty liver). It possibly rises before you get NAFLD as a defense against it too if you're lucky.
What causes fatty liver? Too much energy % from linoleic acid and/or fructose, and not enough choline is the simple formula.
Also, linoleic acid itself contributes carbon to DNL of other lipids, including palmitate and cholesterol.
So the idea that DNL is insignificant in obesity may not be true when at risk of NAFLD, as on SAD, although disruption of hepatic regulation may be more important than the lipid itself.

Jane said...

You could be right about DNL being significant in obesity. It makes a ligand for PPARalpha, which activates transcription of fat-burning genes and prevents fatty liver.

'The nuclear receptor PPARalpha is activated by drugs to treat human disorders of lipid metabolism. Its endogenous ligand is unknown. PPARalpha-dependent gene expression is impaired with inactivation of fatty acid synthase (FAS), suggesting that FAS is involved in generation of a PPARalpha ligand. Here we demonstrate the FAS-dependent presence of a phospholipid bound to PPARalpha isolated from mouse liver. ...Portal vein infusion of [the phospholipid] induced PPARalpha-dependent gene expression and decreased hepatic steatosis. ..'

So it looks like DNL doesn't cause fatty liver, it prevents it. If the same thing happens in adipocytes, it should make you thin, not fat.

Unknown said...


I think your last sentence says it all. We all have to be more explicit when discussing these issues as to whether we are talking about relatively healthy individuals or those in a diseased state. My comment was concerning those not in a diseased or dis-regulated condition. Once that occurs all bets are off imo, and although much is known, the level to which any individual may respond and the severity or the true capability of those responses is likely not truly known.

Anonymous said...

If I may be allowed to add to this interesting discussion again - a relatively low-carbohydrate vegetarian diet has succeeded at keeping me thin (BMI<20) for approximately 10 months. I do not believe it is the vegetarianism that is doing it, but the selection of whole, healthy foodstuffs and minimization of grain, fruit and refined sugar and starch. On days when I don't exercise I voluntarily ensure that I eat less (particularly less protein). On days when I do exercise, I eat more (particularly more protein). Elimination of much of what is contained in the standard American diet, consciousness of net carbohydrate content, regular exercise, and motivation not to eat between meals, snack or overeat at mealtimes has been very helpful. Another key has been standardizing what I eat, tracking its nutritional value, and not varying the menu too much from day-to-day (in fact, I'd say, not varying it much at all, so I have little incentive then to overeat).

Had I continued going the way I was going, I am sure I would have had type 2 diabetes by the time I hit 45. I certainly had a constellation of risk factors indicative of metabolic syndrome. That was 43-45 pounds ago.

The above comments on linoleic acid are interesting to me, as I consume about 50 g per day - somewhat hard for a vegetarian to avoid it. The literature on LA is baffling and contradictory. I hope it is not a 'bad actor' but all I can say is that I do not know. I've seen evidence to support both sides of the argument.

Unknown said...


Congratulations on the weight loss! I'm sure you feel much better!

You do know that the vegetarianism has simply allowed you to cut calories overall?

This subject is so complicated and subjective that nobody truly knows. For instance you are making your workout days the higher protein days, but there's quite a lot of literature suggesting that exercise induced anabolism lasts for up to 24 hours and beyond! Therefore one could make the case that your high protein days should be the following day! In truth it likely doesn't matter enough to be recognizable.

The point of this is we don't know and people all around the planet are trying to split hairs when all we have for a tool is an axe!

What is known works for most when tried. There are outliers and I presume that's what Stephan is trying to figure out with hopes that it will also relate to the large masses (pun intended) who likely would or will respond to the basics that are known.

I'll be willing to bet that the overweight or obese people in the US that don't know what foods they shouldn't overindulge in is small, albeit there are some. However the majority, imo, don't truly understand how many calories they actually eat a day, and often when trying to lose weight think they are doing the right thing. A dichotomy I know, but when you discuss this with them they know by in large what is contributing to the problem. The reasons they can't stop is another issue!

I think we pretty well can agree that's it's a mixture of psychological, and physiological/metabolic issues and the influence of each is different in all people. In the end the education available with regards to diet and nutrition for the public is worse than poor. We continually have so called "experts" discussing foods in isolation as "good or bad", "clean or dirty", "healthy or unhealthy" or whatever label you can come up with! Rarely or ever do you hear discussions of whole diets that are healthy or unhealthy.

Availability and a lack of knowledge, starting at a young age, coupled with a new attitude towards being large is a problem. Maybe we should start with a better understanding at a younger age. But I'll tell you I'm not quite sure if it's my business for those that aren't genetically predisposed. For those people help would be welcome.

Sorry for the rambling!