The “Fat Trap” and biological determinism


As of September 2017, new Sweat Science columns are being published at Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.

- Alex Hutchinson (@sweatscience)


I’ve received a few e-mails asking what I thought of Tara Parker-Pope’s recent New York Times Magazine piece (“The Fat Trap”), which talks about how the body fights off your attempts to make it lose weight. In general, I thought it was a good piece. The basic message I came away with is the same one I hear from people like Yoni Freedhoff: if you want to lose weight — and keep it off — you have to do so using an approach that you’re prepared to maintain for the rest of your life. You can’t go on a diet for six months, lose weight, and then resume your previous diet and lifestyle (or even go halfway back to your previous diet and lifestyle!). Many (perhaps even most) people who are trying to lose weight still see it as a temporary transitional stage. That’s not how the body works, and the more widely that message is spread, the better.

Having said that, a quick note about the apparent “biological determinism” that opposes weight loss. Parker-Pope discusses some of the research by Rupert Leibel’s group at Columbia University, in which subjects are placed on carefully controlled liquid diets to make them gain or lose weight in order to observe what changes take place in their metabolism:

The research shows that the changes that occur after weight loss translate to a huge caloric disadvantage of about 250 to 400 calories… Muscle biopsies taken before, during and after weight loss show that once a person drops weight, their muscle fibers undergo a transformation, making them more like highly efficient “slow twitch” muscle fibers. A result is that after losing weight, your muscles burn 20 to 25 percent fewer calories during everyday activity and moderate aerobic exercise than those of a person who is naturally at the same weight.

I also discuss this research in (plug alert!) my book, Cardio or Weights. And the part Parker-Pope doesn’t mention is that, when you feed people extra calories, exactly the opposite adaptation takes place. In other words, after gaining weight, your muscles burn about 15 percent MORE calories during everyday activity and moderate aerobic exercise. Parker-Pope presents the research as a part of the explanation for why it’s near-impossible to lose weight — but looking at the whole picture, that would mean that it should be impossible to gain weight in the first place! Sure, the barrier is a little bigger when you’re trying to lose weight (20-25% vs. 15%). But the point is, these changes in metabolic efficiency aren’t insurmountable barriers — otherwise no one would ever change weight at all.

23 Replies to “The “Fat Trap” and biological determinism”

  1. Thanks for the link, Todd — I hadn’t seen the new paper yet, and it’s certainly very interesting. Perhaps it’s a topic for a separate post… but my first reaction is that it’s not clear to me why indiscriminate damage to the hypothalamus caused by inflammation would cause the set-point to suddenly decide to defend a weight 20lbs (or whatever) higher than before.

    As an analogy, if I took a hammer to my thermostat, I’d expect it to stop regulating my home’s temperature (or at least regulate it less effectively), and to have an equal chance of moving the temperature up or down. On the other hand, if some undisclosed mechanism raised the thermostat of every house on my block by 10 degrees, I’d surmise that something was resetting the thermostats, not damaging them.

    Of course, that may just be semantics. The research seems like a very interesting avenue, and Guyenet is, as usual, careful not to overstate the findings. They found correlation, not causation — and in this case, it’s not yet clear to me what the causal mechanism would be.

  2. You also point out in Cardio or Weights that the Columbia researchers suggest that more intense exercise might help, since “the biggest change in muscular efficiency was observed at the lowest levels of exercise.” I also wonder what happens if you factor changes in body composition, i.e., if you replace some fat with muscle, even if the muscle is more efficient, you now have more of it, so could that temper the effect?

  3. Now let me get this straight: Bouchard, linked to in Parker-Pope’s article, assumes that genetic disposition leads some people’s bodies to build more FFM without any resistance training or other exercise and it even happens in identical twins?

    Have these results been repeated?

  4. Perhaps most weight loss progammes -all of a sudden no food, lots of running and other stressful activity- are to the body like a simulated famine, thus vindicating the wisdom of buffering fat. I know this is Darwinism of the the-trait-is-there-so-it-must be-useful-so-it’s-there kind, but so are most explanations.

  5. @Penny: Excellent — you not only bought the book, but also read it! 🙂 Yes, my interpretation of that research is that higher intensity exercise may help fight against the efficiency changes, though it’s still speculative. And I think there’s no doubt that changing body composition by adding muscle can help burn more calories — though that’s easier said than done.

  6. @Evilcyber: It doesn’t seem surprising to me, but perhaps I’m missing your point. Bouchard took a bunch of twins and massively overfed them, and noticed that each pair of twins tended to gain weight in similar patterns — my understanding is that this finding has been repeated numerous times. Of the ~20 lbs that the subjects gained, not all of it was fat. Again, this doesn’t seem surprising: in the presence of caloric excess, some people will add muscle just from the activities of day-to-day life. They’re adding twice as much fat at the same time, so it’s not like they’re getting ripped just by overeating!

  7. @Todd: Wouldn’t that mean that once you gain weight, you can’t gain any further weight?

    @RH: Pseudo-Darwinian though it may be, I find that explanation more intuitively plausible. Of course, “intuitively plausible” isn’t always a good proxy for “correct.” 🙂

  8. Alex, I think that Todd is suggesting you are taking your own thermostat analogy too literally. It is logically possible that the hypothalamic “thermostat” works solely by inhibition – that left to its own devices, the “furnace” is always on. In that case, whenever the ability of the thermostat to produce signals is impaired, the temperature will rise, never fall. Obviously, this is an empirical question and your point about the need to prove a causal mechanism is well-taken. But many hormonal systems do work this way. You are wrong to suggest that there should be a presumption that damage to the the hypothalamus is even-handed.

  9. @Phil: Fair point that I can’t presume that damage would be even-handed. But I thought the hypothalamic set-point WAS a two-way regulator: the body defends its set weight from gain OR loss.

    @Todd: On further thought, maybe I’m taking too narrow a view of obesity. A person who is 10 pounds overweight and can’t seem to lose those stubborn pounds may be dealing with a totally different mechanism than a person who is 200 pounds overweight. The former case seems like someone whose set-point is stuck at a higher than desired value; the latter case may well be someone whose set-point is totally broken, whose weight only stops rising because they’re desperately trying to control it. So that would certainly fit with the idea that inflammation in the hypothalamus eventually disrupts weight regulation.

  10. I think I see Phils point. If the set point worked like resting heart rate where the parasympathetic influence through the vagus nerve puts a ‘break’ on the intrinsic heart rate that would be an example. If the parasympathetic influence were to become dysregulated then resting heart rate would only move in one direction (faster resting rate).

  11. @Seth: And if that was the case, then you’d continue gaining weight indefinitely rather than locking into a new, higher set-point — unless the hypothalamus suddenly became undamaged again. The “biological determinism” argument that Tara Parker-Pope wrote about in the NYT assumes that the body’s set-point mechanism is working very well — too well — in obese people. The “inflamed hypothalamus” argument that Stephen Guyenet describes seems (to me, at least) to imply a failure or weakening of that set-point mechanism. As I suggested above, perhaps those two pictures could apply to different types of obesity — moderate and severe, respectively.

  12. Thanks Alex. I may be way off base. I was thinking along the lines of chronic stress causing an imbalance in parasympathic/sympathetic regulation (too much sympathetic influence in what should be a rest state). So that would be a change in appropriate regulation rather than a total parasympthetic withdrawl.

    It may not be an appropriate analogy. The way I look at things however things are not working ‘well’ when human bodies get out of balance. The ‘set point’ in anorexia or extreme obesity is not where it should be to help a person regain balance, it has seemingly lost it’s regulation in a vicious cycle. I wonder if it would be interesting to see what is going on in the hypothalmus with anorexia patients and how it relates to what is seen in obesity. I know there has been research relating abnormal HPA function with increased activity in dopamine recepters in anorexic patients.

  13. With a quick google i found this:

    ‘Two regions of the hypothalamus, the ventromedial hypothalamus and the lateral hypothalamus, have been shown to effect feeding behavior. In particular, the ventromedial hypothalamus is thought to contain a satiety center, while the lateral hypothalamus is thought to contain an eating center. This hypothesis comes from animal experiments in which these areas are selectively stimulated or damaged. Thus, the ventromedial hypothalamus is thought to be the satiety center because stimulation of this area suppresses eating, while damage to this area results in obesity. The opposite effects are seen following stimulation and damage to the lateral hypothalamus, thus suggesting that it may be responsible for promoting eating behavior. The mechanisms underlying these regulatory processes seem to involve the activity of the neurotransmitters norepinephrine, serotonin, and dopamine. It is suggested that further investigation into the role of these hypothalamic regions and the neurotransmitter pathways within them may provide important information about the biochemical and neurological basis of eating disorders (2)’


    The source it cites ( seems dead so I can’t vouch for the info but it seems relevant and interesting.

  14. “A result is that after losing weight, your muscles burn 20 to 25 percent fewer calories during everyday activity and moderate Aerobic exercise”

    This would lead me to believe that one must do strength/interval workouts to force the body to maintain type IIa/b muscle fibers if not the body will convert to type I.

  15. Alex,

    Whatever is really happening in obesity it’s probably way more complicated than hitting a thermostat with a hammer. More like disturbing an ecosystem or a market maybe. Either way, it does seem like something is “broken” in obese people, whatever that something may be.

  16. @alex: Yep, of course. But Bouchard claims he had all those parameters under control (amount of physical exercise etc.) and yet there was a difference.

    That struck me as rather remarkable.

  17. @Evilcyber: I’m still not sure whether I’m missing something. Are you really suggesting that, if you gave identical workouts to a group of people for six months, you’d expect them to all put on exactly the same amount of muscle?!

  18. ” Muscle biopsies taken before, during and after weight loss show that once a person drops weight, their muscle fibers undergo a transformation, making them more like highly efficient “slow twitch” muscle fibers. A result is that after losing weight, your muscles burn 20 to 25 percent fewer calories during everyday activity and moderate aerobic exercise than those of a person who is naturally at the same weight.”

    I’ve read a few papers during the last few months on how resistance exercise, activation of the P3IK-AKT pathway (stimulated through insulin signaling), and the ingestion of branched-chain amino acids stimulate mTOR, which leads to an increase net muscle protein synthesis by deactivating efE4, the ribosomal protein that binds the 5′ methylguanine cap during translation and activates p70 S6 kinase. Resistance exercise primarily works by stimulating erk1/2 via the MAP kinase pathway. Erk1/2 can stimulate mTOR activity by directly phosphorylating TSC2, a GTPase activating protein whose substrate is the Rheb GTPase (Rab homologue enriched in brain), deactivating it. When the GTP bound form of Rheb is bound to mTOR, this stimulates mTOR’s serine/theorine kinase activate; thus by inhibiting TSC2, Erk phosphorylation inhibits an inhibitor of mTOR. Another downstream target of Erk1/2 is p90 RSK (which, like p70 S61 kinase, also phosphorylates ribosomal protein S6, stimulating more translation) and, like its upstream kinase erk1/2, it can inhibit TSC.

    I wonder if activating Erk1/2 through resistance training would also discourage the transition of type II (fast-twitch muscle fibers) to type I (slow twitch, oxidative fibers) , in addition to the effects I briefly elaborated on hypertrophy . Also, I would presume that caloric restriction (inherent in any weight loss effort) would increase AMPK activity, which phosphorylates TSC1, by unlike erk1/2 and p90 rsk, AMPK phosphorylation stimulates its activity, leading to less muscle hypertrophy.

  19. I think what Bouchard showed was that the variation between twins is way higher than within twins . The weight gain by identical twins were a lot similar than weight gain by other twins. This was shown for weight loss too. It basically tells us that people can eat the same food and gan different weight.

    From what I have read, it think it is pretty clear that there strong biological basis for obesity. The tiwn studies shows the heritablty of obesity to be second to height.

  20. Hi Alex,

    The following blog: (see also part2)

    has two relevant posts I thought you might be interested in. A recent study as shed some light on how neurons within the hypothalamus function both to stimulate and reduce the hunger response. Might provide as a source for speculation on how a particular type of disruption to the feedback mechanism could possibly escalate always in the same diretion (more hunger).

    Also an interesting finding on ‘brown fat’ in the NY times today. I figure you will probably see that one as well.


Comments are closed.