When last I posted, oh so long ago, I promised next to discuss the food reward/palatability hypothesis of obesity and why I find it so uncompelling and more than a little bit disheartening. Why, in effect, I think it is the kind of bad science that begs to be challenged, as I did, when it is presented in a public forum, as Dr. Stephan Guyenet of Whole Health Source did at the Ancestral Health Symposium back in August. This is the first of five posts to address this and I promise (really) that it will be days between posts, not months.
Here’s how Dr. Guyenet describes his hypothesis in his sixth of an increasing number of blog posts on food reward:
…The evidence as a whole shows that chronic consumption of foods with an excessive reward value causes abnormalities in parts of the brain that regulate body fatness, metabolism and reward/motivation. This can lead to weight gain and metabolic problems, and favor addictive and compulsive relationships to food and other things. The combination of readily accessible, cheap, high-reward food, and stressful lifestyles that drive us to eat it, is probably a major contributor to overweight, obesity, diabetes and perhaps other health problems in affluent nations.
What makes this hypothesis novel is that the ability of a particular food to make us fat — what makes a food fattening, in effect — is mediated, according to Dr. Guyenet, through its ability to stimulate neurochemical effects in the brain.
So why do I find this intriguing idea objectionable, so objectionable, in fact, that I was willing to stand up at the AHS and make a bit of spectacle of myself? Well, for starters, the food reward/palatability hypothesis can’t explain the kinds of observations about obesity and weight regulation that I’ve been arguing in my books must be explained by any viable hypothesis of why we get fat. This is the primary issue, obviously, as a hypothesis that can’t explain the necessary observations is a failed hypothesis. And we’ll get to this issue in the posts to come.
The lesser reason for my objections, and the subject of this post, is that this hypothesis seems to be yet another thinly-veiled variation on the energy balance paradigm of obesity — calories-in, calories-out — and I’ve been arguing that this particular paradigm is fatally misguided in a variety of different ways. I believe that obesity researchers will never make meaningful progress until they rid themselves of this belief system, and that so long as they hold onto it they’ll be effectively blaming the obese for the problem of their obesity. Hence, finding yet a new way to embrace it, or popularizing an old way, as Dr. Guyenet has, seems to be a step in precisely the wrong direction.
Ultimately, this comes down to what I consider the two competing obesity paradigms or belief systems.
Competing paradigms: 1. body rules
One is a body-rules paradigm and I’ve been arguing that it’s the only one that can do a reasonable job of explaining all the meaningful observations. Changes in the hormonal and enzymatic regulation of fat metabolism—in the regulation of fat storage and oxidation—drive changes in adiposity; and changes in adiposity drive compensatory changes in intake and expenditure.
Here, the body is running things. Indeed the organ in control may be the fat tissue itself in concert with the liver. The University of Vienna endocrinologist/geneticist Julius Bauer described this fat-rules concept back in 1929 by saying that the fat tissue of someone who’s obese (what he called “abnormal lipophilic tissue) “maintains its stock, and may increase it independent of the requirements of the organism. A sort of anarchy exists; the adipose tissue lives for itself and does not fit into the precisely regulated management of the whole organism.”
In this scenario, the brain plays no more role in regulating the growth of the fat tissue than it would regulating the growth of any tissue — a tumor, for instance, which is a metaphor that Bauer uses, or the growth of fat tissue in breasts and hips and butt when a young girl goes through puberty, or the fat put on by a mother-to-be when she is pregnant, another Bauer metaphor, or just the growth of lean and fat tissue in children as they develop and mature.
In all these cases, the body is doing its own thing, more or less independent of the brain’s input, even though the brain, in the case of growing children, is the source of the growth hormone that is driving the growth. More energy has to be taken in than expended while growth is occurring (at least by the tissue itself, if not the whole body), but that “positive energy balance” — calories in greater than calories out — is a compensatory effect of the growth; it’s not a cause. The brain does what it always does, which is reacting and modulating homeostasis in response to environmental signals. But it’s a secondary organ in this sense.
In this paradigm, specific foods are fattening because they induce metabolic and hormonal responses in the body — in the periphery, as its known in the lingo — that in turn induce fat cells to accumulate fat. The brain has little say in the matter. As I describe it in my books (and as it was described to me by the few researchers today who hold this view), we don’t get fat because we overeat; we overeat because we’re getting fat.
If we come to crave specific foods or find them more rewarding than others and these foods are fattening — not necessarily the same thing — then we likely do so because of their physiologic effects, the changes in the hormonal milieu of the body caused by the consumption of these foods. Foods that make us fat come to be foods we crave, because these foods are also the foods that make us hungry. So how much we might like a food or how rewarding we might find it or how much we might crave it are primarily consequences of the peripheral metabolic/hormonal effects.
In this paradigm, meals stimulate hormonal responses—insulin, in particular, either in the short term (glucose) or the long term (fructose)— and this in turn directly influences both the storage of fat and the oxidation of fatty acids elsewhere in the body. The balance of forces working on the fat tissue, a dynamic equilibrium, ultimately determines how much fat we accumulate. As fat accumulation increases, so does secretion of leptin and other adipokines that work in a variety of feedback loops in the brain and the body (as does insulin itself) to ideally prevent excessive fat accumulation.
This body-rules hypothesis, of course, isn’t the conventional wisdom. If it were, I would be doing something else with my career.
Competing paradigms: 2. brain rules
The conventional wisdom is that we get fat because we take in more calories than we expend. Simple enough. We get fat because we overeat, not the other way around. Changes in energy balance—calories-in minus calories-out—drive changes in adiposity, in how much fat we carry around in our fat cells.
Ultimately, as I discuss in Why We Get Fat, this is a brain-rules paradigm. After all, both the components of overeating — eating too much, aka gluttony, or moving too little, aka sloth — are both behaviors and in this paradigm behaviors are psychological phenomena not physiological.
Researchers who live in this paradigm are invariably trying to discover what’s wrong with our brains or the signaling to our brains that cause this particularly cherished organ (what Woody Allen memorably described as his “second favorite organ”) to screw up. Why can’t the brains of people who become obese or overweight get the energy balance thing right? Or why do these brains effectively desire more fat on the body than is healthy? Why do they set the “set point” of adiposity too high? The problem with people who get obese is in their brains, not their bodies (even though the excess fat is in the body).
In this paradigm, the fat tissue is more or less passively following the brain’s lead. Meals will stimulate a variety of gut hormones and fat-derived hormones like leptin, grhelin, etc., and these in turn signal the brain whether to continue eating or not, whether to act on hunger or satiation, and perhaps how much energy to drive the body to expend. But ultimately the deciding factor on whether a person gets fatter or leaner is the balance of energy expended or consumed—calories-in minus calories-out—and this is regulated by the brain. The fat tissue’s say in the matter is mostly relegated to its secretion of leptin and other adipokines, as these fat-derived hormones are known, that might in turn work in the brain to curb hunger or maybe work in the body to increase expenditure. Energy balance, though, drives adiposity and energy balance (energy homeostasis, as its often called) is ultimately a brain thing not a body thing.
The twist that the food reward/palatability hypothesis of obesity gives to this energy balance paradigm is to make the meaningful effect of a fattening food that of changing the neurochemical balance in the brain. This in turn changes set point and adiposity increases (or decreases) in response. This scenario can appear to be one that rejects, as I do, the simplistic calories-in-calories-out idea, but I’m going to argue that this appearance is an illusion. And that, as I said, is one reason why the hypothesis irks me so.
Here’s one way to look at these two competing paradigms, now with the food reward/palatability hypothesis of obesity standing in for the conventional wisdom:
Here’s the fat-rules paradigm: We’re not getting fat because we are overeating; we are overeating because we’re getting fat—because a hormonal disorder, caused by the consumption of certain foods (refined grains, easily-digestible starches and sugars) drives us to accumulate fat in our fat tissue.
And here’s the food reward/palatability hypothesis, a brain-rules hypothesis: We’re not getting fat because we are overeating; we are overeating because our brains are driving us to get fat—because a neurochemical disorder, caused by the consumption of certain foods (refined grains and sugars, as well as fats and salty foods and industrially processed foods of other types as well), in turn drives us to accumulate fat in our fat tissue.
So you can see how confusing this can be, how similar these two hypotheses can appear. But the question ultimately is how does the brain go about increasing fat accumulation in the food reward/palatability hypothesis. What do these neurochemical changes do to the brain and then what does the brain do that ultimately leads to more fat in the fat cells, which are, after all, mostly half-a-body or so away? What’s ultimately driving the fat?
Here’s where the food reward/palatability hypothesis seems to collapse back down to just another failed twist on calories-in-calories-out. It’s yet another attempt to explain why fat people get fat, and why it’s caused by eating too much (or exercising too little), without explicitly blaming obese or overweight individuals for not being able to control their behaviors (as we lean people do). Chapter 7 in Why We Get Fat goes into this problem in detail.
(A caveat here: I’m willing to be convinced that this hypothesis is not an energy balance hypothesis, and so not a brain-rules-and-the-fat-passively-follows notion, but so far I haven’t seen an argument that’s convincing. I confess, though, that I find the hypothesis surprisingly difficult to understand, which suggests that either I’m losing my intellectual facilities in my dotage — always possible — or that I’m so blinded by my love of my own hypothesis that I refuse to understand it—at least equally likely — or maybe, just maybe, that Dr. Guyenet and other proponents of the hypothesis don’t really understand it either.)
The food reward/palatability hypothesis: looking for a mechanism
As Dr. Guyenet says, as I quoted above, the chain of causality in this hypothesis goes like this: “abnormalities in parts of the brain that regulate body fatness, metabolism and reward/motivation… can lead to weight gain and metabolic problems…” But he doesn’t say how. How do abnormalities in parts of the brain lead to weight gain? The answer seems to be that the brain raises our adiposity set point, but then how does that actually make us fatter? Now we have a change in the brain — set point has gone up — but we still have to explain the change in the body — adiposity increases. How does that happen?
And the answer is that the brain affects this change in fat mass, as proponents of the hypothesis seem to see it (or seem to see it most often), by either increasing food consumption or decreasing expenditure. So the brain is regulating the fat mass not by regulating fat accumulation directly, but by making us eat more and maybe expend less — calories-in-calories-out. And fat people get fat not because their fat tissue is living for itself, as Bauer put it, and increasing its stock of fat, independent of the nutritional state of the organism, but because fat people either respond to rewarding foods differently than lean people do or because they can’t resist this drive to eat more food once their set point goes up. Lean people either can resist the urge, or the rewarding foods don’t have the same neurochemical effects in their brains. Either way, the brains of lean people are more resistant to fattening foods (those with high food reward/palatability value) than the brains of fat people, and the body, in either case, is kind of irrelevant.
Now, the brain could regulate fat mass directly by increasing, for instance, insulin secretion via increased stimulation of the vagus nerve, but Dr. Guyenet would like to dismiss the role of insulin in excess fat accumulation. (This is another post I’ve promised in my last post and one that might come along, well, maybe in a month or two, after we’re done with these food reward posts). As a result, this particular mechanism is not apparently part of the hypothesis.
When I asked Dr. Guyenet, shortly after my last post, why he believed so strongly that the food reward hypothesis was not an energy balance idea, he suggested that I could significantly improve my understanding of this issue by reading his current mentor Michael Schwartz’s 2006 review article in Nature – “Central nervous system control of food intake and body weight.”
A lengthy but not necessarily irrelevant digression
Curiously enough, I had a run-in with Professor Schwartz himself over precisely this issue back in March 2005 when I was reporting Good Calories, Bad Calories. Schwartz built his reputation in obesity research in part by demonstrating that insulin in the brain suppresses food intake – or at least that it does when it is injected directly into the cerebral spinal fluid of baboons. He then argued that this is the hormone’s primary role; Dr. Guyenet is now making very similar arguments. In their opinion, what happens in the brain trumps what happens in the body. Schwartz has actually suggested that one reason for the obesity epidemic may be that we’re not secreting enough insulin because we’re not eating enough carbohydrates. Dr. Guyenet does not go that far.
I argued in an e-mail exchange with Schwartz that because insulin is secreted in the periphery, and that because it can be shown in both humans and animals that increasing insulin in the periphery increases fat accumulation, this would seem to be the primary effect. The appetite-suppression effect in the brain, therefore, would be a secondary effect—a negative feedback loop of the kind that you would expect to see in a homeostatic system.
Schwartz took exception to being questioned by a journalist and he argued that the role of journalists in this field was that of ”helping to debunk profit-motivated, unscientific views that are so prevalent in our culture.” I took exception to being told what my journalistic role should be and replied that my role was to help debunk unscientific views prevalent in our culture, regardless of their institutional provenance or apparent motivation. It took us several e-mails back and forth to achieve a peaceful co-existence again.
Back on track: the mechanism of food reward/palatability
The Schwartz article that Dr. Guyenet suggest I study, though, makes it relatively clear that the food reward/palatability hypothesis of obesity is indeed firmly entrenched in the energy balance paradigm. (I pointed this out to Dr. Guyenet in an e-mail, and he said he would get back to me on it. Apparently other concerns interceded on his side, though, and so I write this without benefit of his clarification.) “Through a process known as energy homeostasis,” Schwarz and his co-author write, “food intake is adjusted over time so as to promote stability in the amount of body fuel stored as fat.” And energy homeostasis, they explain, is regulated via a negative feedback model:
Introduced more than 50 years ago, the `adiposity negative-feedback’ model of energy homeostasis is founded on the premise that circulating signals inform the brain of changes in body fat mass and that in response to this input, the brain mounts adaptive adjustments of energy balance to stabilize fat stores.
So the brain adjusts energy balance (intake and expenditure) to stabilize fat stores; the brain rules, we eat more or expend less, and the fat tissue inflates or deflates accordingly. And this is why Dr. Guyenet and others can write thousands of words about food reward and palatability without ever actually discussing the hormonal/enzymatic regulation of the fat tissue itself, which is, after all, the ultimate organ of interest, and without discussing the hormonal/metabolic regulation of fatty acid oxidation in the lean tissue, when it’s the fatty acids that are ultimately being stored.
The alternative would be to suggest that the brain adjusts fat stores directly via hormonal and central nervous system mechanisms, and this leads to compensatory changes in energy balance. But then, as I said, we would have to discuss the direct peripheral effect of insulin on fat storage and fatty acid oxidation, and that’s not allowed.
“Obesity, by definition, results from ingesting calories in excess of ongoing requirements,” Schwartz and his co-author write, and I’m arguing that any researcher who makes this statement for any reason other than to make the point that it’s meaningless and misleading is living and working in an energy balance paradigm — calories-in, calories-out.
And because I think the energy balance paradigm of obesity is more or less the root of all scientific evil in this business — not just because it’s taken obesity researchers down a century-long blind alley, but because of its implications that fat people just can’t control their urges the way lean people do — I find Dr. Guyenet’s promotion of this hypothesis and its acceptance, limited as it may be, in the paleo and low-carb blogospheres to be very disheartening. I could be wrong (of course) about the scientific bankruptcy of this hypothesis, but I’m going to argue (of course) that I’m not.
In my next post—just a few days from now, I promise (barring extenuating circumstances like the Hayward fault underlying our neighborhood deciding to go off with a magnitude significantly higher than the 3 to 4s we’ve been getting for the past two weeks) — we’ll continue this discussion by looking at the other major limitations of this hypothesis, beginning with some recent observations that it can’t seem to explain.