By MARILYNN LARKIN

Dr. Barry E. Levin, a neurologist at New Jersey Medical School, has been investigating the brain’s role in obesity for more than 20 years. His recent findings suggest that exercise early in life may prevent obesity, even in people who have a genetic propensity to gain weight.

Q. What drives people to overeat?

A. Balancing food intake and energy expenditure is a complex process that involves two systems in the brain: a need-based system that triggers hunger when the body requires energy, and a reward-based system that triggers the desire to eat. The two sites are close to each other anatomically and work like parts of a machine.

The need-based system involves metabolism. It’s the part that monitors calories and receives signals from the body — for example, shifting levels of glucose or the hormones insulin and leptin — that tell you you’ve eaten enough and it’s time to stop. If you’re genetically prone to becoming obese, you may not be very sensitive to those signals, and you’ll eat more before stopping than someone who is not obesity-prone.

Everyone’s system strives to maintain a constant weight, or set point. The set point can be shifted upward, but only over time. If you overeat during a single meal — say, a big Thanksgiving dinner — your body goes into overdrive to try to get rid of those excess calories. But if you gain weight gradually, your body eventually will become accustomed to a higher weight and work to maintain it.

Q. What about the brain’s reward system?

A. The reward-based system, the part that responds to the tastes and textures of your favorite foods, and the joys of dining out with family and friends, is linked to many of the same metabolic pathways as the need-based system. But it plays by a different set of rules. We saw an example of its power in a series of experiments with obesity-prone animals who love a particular chocolate liquid supplement.

When the supplement was available, the animals ignored the high-fat food that would normally have attracted them and ate almost all their calories from the supplement. But when the supplement was taken away — that is, the palatability of their diet changed — they cut their food intake by 50 percent. Within two to three weeks, their weight was back where it would have been if they had never eaten the supplement.

When the supplement was reintroduced, however, the animals again ate with abandon, and their weight went right back to where it would have been if they’d never been off it. So this reward system seems to override everything, even normal signals of hunger and satiety. It drives many people to overeat and probably is responsible for much of the increase in obesity we’ve seen over the past few decades.

Q. What role do genes play in obesity?

A. Genes are responsible for 60 to 70 percent of obesity, in that they can program you to be obesity-prone. Your set point is genotype-specific, meaning it’s tied to your genetic template, and it’s very difficult to change. In fact, once an obesity-prone individual actually becomes obese, it’s nearly impossible to return to a normal weight. So although many people believe that individuals who are overweight simply lack the willpower to stop overeating, in fact it’s much more complicated. If someone is obesity-prone, many times they are also highly sensitive to cues in the environment, such as the smell or taste of a favorite food.

Q. Why don’t diets help?

A. One problem is that when you lose weight on a diet, the amount of energy you expend at rest — part of the need-based system — goes down. Although dieting affects both the need and reward systems, it’s need that can’t be overcome. Built into every one of us, fat or thin, is a metabolic system that drives us to eat when the brain perceives that we don’t have enough energy stores. It’s this perception that can be skewed in an obesity-prone person, whose brain is less sensitive to signals to stop eating.

When you fast, even for as little as 12 hours, levels of the hormone leptin drop like a shot, which is a way for the brain to know the body is hungry. When leptin levels go back up, the brain says, “I’m full.” But obesity-prone people are less sensitive to high levels, and don’t get the message to stop until levels are really, really high.

We saw this when we compared obesity-prone animals to obesity-resistant animals. We fed them both a high-fat diet. Both took in more calories than usual, but after two or three days, something clicked in the brains of the obesity-resistant animals, and they lowered their caloric intake back down to where it was when they were on a low-fat, lower-calorie diet. But it took the obesity-prone animals three or four weeks to make that correction, even though their leptin levels had gone way up.

People probably fall somewhere in between; they don’t overeat quite as much, and a good part of the calories that they put on contribute to their need-based increase in weight, which is what you can’t overcome by changing palatability. In fact, I believe a lot of fad diets work initially, at least, by being boring. People lose interest in eating. But eventually, they start craving their favorite foods, and they can’t stay on the diet anymore.

Q. Does exercise really make a difference?

A. A large group of people in the National Weight Control Registry report doing very heavy amounts of exercise as part of their weight management strategy. And we know that exercise helps overcome the body’s natural tendency to lower its resting energy expenditure in response to weight loss. But for most people, exercise in and of itself isn’t the answer.

What we are finding is that exercise done very early — immediately after weaning in animals, and probably anywhere from age 2 to 8 in humans — may help prevent someone with an obesity-prone template from becoming fat. We did an experiment in which a running wheel was put in a cage of animals who enjoy using it. They ran on it like crazy, and we found that even animals with a propensity to obesity did not gain weight, even after the wheel was taken away. The effect was incredibly dramatic.

We had never been able to change the set point of the obesity-prone before, and we suspected that exercise done during this early developmental period somehow made their brains more sensitive to leptin signals. We don’t know if the effect is permanent, but it certainly is long-lasting. And it’s encouraging because it shows that there may be a way to change the obesity genotype with an early intervention.

Q. Industry has spent a fortune trying to develop effective weight-loss drugs. Why don’t we have one yet?

A. First, the brain is effectively wired to protect itself. We need to eat to survive, and we have redundant pathways developed over eons that keep us wanting to eat, pushing us to keep our energy systems filled. If you try to interfere with one, others take over.

But a bigger problem is that the receptors these weight-loss drugs act on don’t just regulate food intake; they’re tied to a lot of other systems. That’s true of rimonabant, a weight-loss drug that the F.D.A. has refused to approve. It works on the brain’s cannaboid receptors, which are involved in eating and the reward system, but also in anxiety, depression and mood.

And this ties into the biggest problem in treating obesity, which is that people are impatient. If someone responds to such a drug, he may lose a pound or so a week. And no matter what the doctor says, he’ll become impatient and take more pills. This may speed up weight loss, but because the drug is also acting on other brain centers, the increased dosage can also induce depression, anxiety, psychosis and suicide.

We have plenty of targets for new drugs. What we need is something more like a smart bomb that zeroes in on a weight-loss target without affecting anything else. Some potential drugs are in the wings, but it will take years before they can be tested in humans. And we’ll probably need combinations of drugs that target systems both in the brain and in the body, because one drug targeting one system probably won’t produce more than a 10 percent weight reduction, which is not enough for someone who weighs 300 pounds.

Unfortunately, for now and the foreseeable future, the best treatment for obesity is prevention.