Monday, October 1, 2012

Shangri-La Diet : Results for Third Month : Fail?

Now I'm confused. For this last month I tried modifying this by drinking water just after the oil, so it didn't leave any aftertaste.

As far as I could tell, this worked a treat, I had practically no appetite, and I got thinner. This despite doing literally no exercise this month as I've damaged my knee. I have been smoking quite a lot though, since I took my narrowboat cruising and steering a boat is much more fun with a cigar in one hand.

About a week ago I noticed that my belt was beginning to need to go in a notch, and I figured by the end of the month it would be comfortably on notch four.

But it isn't. In fact I seem to have swelled a bit. But my appetite is not back. In fact I'm finding it a bit difficult to believe that I'm eating anywhere near the number of calories I need to maintain my current weight.

Still, updating beliefs for notch 3/appetite loss:

H 77: W 10: S 13

likelihoods of observed results (clearly a win for helplessness and a defeat for willpower):

H 50/203 : W 5/134 : S 25/134

posteriors, rounded

H87 : W2 : S11

So the usual willpower theory of weight control, having made consistently inaccurate predictions as to the effect of deliberately eating extra calories, is getting annihilated as far as I'm concerned. But helplessness is slowly edging out Shangri-La, since although my belt measurement has gone down over the last three months, it's been stable twice and only dropped once.

I'll keep doing this for at least another month, and I'll keep the same protocol and predictions as last month.


  1. A problem with the Shangri-La Diet is that there's no explanation why tasty foods should raise the calorie "set-point".

    I can offer a stab at an explanation. It is that people overeat in order to get "high" and they do this for emotional reasons. Like the stereotypical sitcom girl who eats a whole pot of Ben and Jerry's in order to cope with being dumped by her boyfriend. Getting high seems to restore the flow of thoughts and one's sense of autonomy in the face of anxiety and pessimism.

    In the same way that an alcoholic's tolerance for alcohol increases, so the overeater will require more and more food in order to get the same "high".

    The way people get high is by deliberately damaging the body in some way. The brain interprets this as danger and releases natural chemicals which fuel the fight-or-flight response (which manifests in the modern human as racing thoughts and bradycardia, although, more primitively perhaps, some people do jump about on the dance floor).

    Here’s an example of a new painkiller where the connection with damaging the body is more obvious:

    Although reducing the set-point may or may not help with one's weight, the hidden motivation for overeating has not been addressed and so the problem will remain and may pop up in other behaviours (like smoking more cigarettes, for example). Otherwise, in the absence of deeper change, my guess is that the dieter will eventually re-learn to crave more food.

    -- Tom

    1. Tom, Seth's got various experiments backing that effect up in rats I think. Does your explanation apply to them as well?

    2. From what I recall of Seth he's a interesting and slightly heroic gentleman, but like many scientists he is a victim of empiricism. The truth is that experiment can only be used to choose between explanations and not to generate them.

      Humans are very different from rats: for one thing rats presumably don't have an internal monologue running 24/7 that demands stimulation ('demands' because we tend to identify with it -- although I note from some fascinating entries of yours that you can perceive it. From what I have learnt this an important step in self-realisation including, crucially, freedom from obsessions and addictions!)

  2. Oops, I meant "tachycardia". There's a price to be paid for choosing pompous medical terminology...

  3. Just spotted on Less Wrong (Firewalling the Optimal from the Rational): Eliezer remarks in passing that Shangri-La "led me to lose twenty pounds that I've mostly kept off, and then it mysteriously stopped working".

    1. Yes, in fact I first heard of it via Eliezer's post:(

      It sounded wacky, believable and easy to test so I decided to have a go at it and try to keep my beliefs straight using his methods.

      It's been very educational! I think I've made just about as much of a mess of it as it's possible to make, and that's been a great learning experience.

      Weirdly, in that post, Eliezer claimed that it hadn't worked for him. In the post you reference, he's claiming that he dropped 20lbs (which is an absolute fuckload!, about 1/9th of my total weight), and that it didn't come back.

      I seem to have thinned out a bit too on balance, but that could perfectly well be a side effect of paying attention to what I eat, even though I've been deliberately trying to eat exactly what I want.