Being "scientific" about exercise and weight control

I was looking for references to research on the relationship of appetite and exercise and I came across an interesting blog post that inspired some thoughts. I see a lot of people trying to interpret research results, which in general I like to see. But people also seem to commonly try to draw advice for themselves from selectively chosen studies, and that's the part that I wanted to comment on.

If you want to know how exercise effects your body, and you have confidence in science, try doing it and measure the effects. Make yourself the study population rather than trying to identify (or not) with the population in the latest study. Or, if instead you want to understand the research, get the background for the program and read the literature directly.

People tend to treat journalist's reports of research it as if it were "scientific advice" and then when it is nothing of the kind (which is generally the case), the researchers are accused of being hopelessly obscure, getting paid for results, or being contradictory about their advice. What a bum rap. That's not the point of research at all.

Research is most *interesting* when it contains contradictions and gets *boring* when the results become consistent. Sure, you want to prove your new hypothesis is right, but if it were easy to demonstrate someone would already have demonstrated it. You expect things to be fuzzy in a new area of research where there are interesting new hypotheses to test.You can almost never generallize "laws" (or advice) from a single study, nor should you expect to.

Conclusions vary even in very similar studies either because of the way the question was asked or because they didn't identify the right variables in the right way yet to address the question. That's assuming they're even studying something that can be characterized in a law-like way, which isn't always the case.

Sometimes individual differences outweigh generalized laws, even when there *are* underlying laws (this is a common finding in social psychology research). More to the point, this seems to be the case somewhat with nutrition as well. Conflicting results are a symptom of a fertile research topic.

IMO, if you want to draw good advice from research, stop jumping on every new study that comes out just because it seems to say something you'd like to hear, and learn to follow the *trends* in the research programs. That means reading technical books and journals rather than sound bites and blogs, and looking for commonalities rather than just reports that stand out because they are novel. But that's more work than you probably need to do.

There are a lot of people that have been spending decades researching obesity and there are some very useful conclusions. But you still have to do the work to see what works best for you.

I think there are many, many fat people out there exercising like crazy and not reaching their goals because they keep looking for magic bullets or excuses instead of starting with sound principles and then experimenting carefully to see what is working for them and what isn't.

I've noticed that there are also a number of chubby runners out there who stay chubby because they take in more than they put out, which I suspect is due to the putative appetite effect. It could well be that they would benefit by changing their training to briefer, higher intensity work, so that the appetite effect would lessen and the metabolic increase last longer. But that's speculation at this point and represents very hard work and many people would give up on such a regimen if they're used to jogging and elliptical machines and such. So they have to experiment to see what's right for them.

Well, anyway, personal experimentation (and brief high intensity training) is an approach that has always worked very well for me. Just remember, people tend to distort their own perception of what they take in and what they put out (*that's* a pretty well validated research result across many studies). So when you measure your compliance and your results, be sure to be specific and use accurate measurements and don't rely on memory and estimation.