Friday, May 26, 2006

The Shangri-La Diet

In the mail: The Shangri-La Diet, by Seth Roberts. Seth Roberts is a psychology professor at Berkeley who has gotten a lot of publicity for his attempts at self-experimentation. His most famous experiment involved figuring out how to limit his appetite by changing his diet -- which led to what he calls the Shangri-La Diet.

The concept is pretty simple: 1) Your body has a "set point" at any given time -- i.e., a given weight that the body "wants" to reach. If the set point is high, then you are hungrier. And vice versa. 2) The set point (and hence appetite, and hence weight gain or loss) rises whenever your body thinks that there is lots of flavorful food around. And it goes down when your body thinks that there is less flavorful food (i.e., a lack of flavorful food tells one's body that you are having to resort to undesirable food, which means starvation conditions, which tells your body to lower the appetite).

And finally: 3) The way to trick your body into thinking that there is less flavorful food around is to consume a few hundred calories a day from two sources that have little flavor: Oil, or sugar water. The calories here are essential: You can't drink saccharine water, because your body won't "think" that the only source of real calories is unflavorful food, and then it won't lower your set point.

Needless to say, the idea that you can lower your appetite and lose weight by drinking sugar water or oil is somewhat weird and controversial. The typical thing to do at this point would be to report on the results of trying it myself. But I don't need to lose any weight, and don't really care to try. So I'll point out three sources of support that Roberts does -- or doesn't -- bring to bear.

First, Roberts does cite a lot of studies and experiments to support the notion that 1) our bodies regulate weight via a set point, and 2) it is possible to regulate this set point via eating either flavorful or bland foods. Fair enough.

Second, Roberts cites a bunch of anecdotes from people -- often commenters on various blogs -- who claim that they have lost weight or found their appetite suppressed on the diet. I'm distinctly less impressed with this source of evidence. Even assuming that all of the commenters are telling the truth, there are two problems: Some of the results may be psychosomatic, and some of the claimed results simply aren't all that impressive. Roberts does report some anecdotes of people who have lost more than 20 pounds. But he also includes a lot of anecdotes like this:

I've lost about 3 pounds in ten days. Not earth shattering, but slow and steady is good I think.

This sort of comment is meaningless. I myself lost 6 pounds the other day in under an hour-and-a-half. It's true. But that was only because I went for a 9-mile run in the middle of the day. I gained all the weight back once I drank a lot of water to replace the sweat that I had lost. The point is, your weight can fluctuate widely depending on how much water you have drunk, what time of day it is, whether you have been sweating, whether you just ate a large meal (or did the opposite). All of which is to say that a loss of three pounds in ten days doesn't necessarily mean anything.

Third, the best source of evidence would be something that Roberts doesn't have: A double-blind experiment where a large sample of people is randomly divided into two groups, one of which is assigned to drink genuine sugar water or oil, and the other of which drinks a calorie-free substitute for the above; and then researchers at the end would determine which group of people lost more weight. (Incidentally, this would probably be the first experiment in history where researchers had to use a placebo for sugar water.)

I guess we'll have to wait for that experiment. For now, Roberts' book is flawed but very intriguing.


Blogger Paul Gowder said...

Worse, it's the only diet that actually relies on choosing the foods specifically because they taste bad! What's next? The hair shirt and flogging diet? :-)

10:16 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home