The limits of exercise (hat tip: MR):

For instance, it seems obvious that increasing energy output—by walking to school, for instance, or starting an intense gym program—will help decrease obesity. Unfortunately, another study points out that what's obvious isn't necessarily true.

T.J. Wilkin and his colleagues at the Peninsula Medical School in Devon, England, looked at three groups of English and Scottish children. They measured physical activity using accelerometers, devices that record duration and intensity of movement 600 times a minute....

Wilkins and his team found that every child has his or her own very consistent daily level of activity. It remains the same on weekends and weekdays; it's not affected by school physical education, or by whether the child walks or drives to school, or how much time he spends awake or in front of a television. We don't know what determines this intrinsic level of activity. But engineering the environment to make available or even to require more activity will apparently have little impact on children whose nature is to be inactive.

The study's subjects' ages were five to nine years. Being the oldest of five siblings, the results don't surprise me. I'm much more interested in how much this homeostatic effect tapers off with age. Fortunately, the paper mentions that some of its subjects are part of an ongoing longitudinal study that will address this.