On a flight from Philadelphia to Chicago that one of us took some time ago, a full lunch was served at 4:00 p.m. Six of the nine passengers in sight ate all of it. Simply put, it was there. And in a study conducted with people who have amnesia, and who therefore cannot remember that they have just eaten, the subjects consumed second and even third lunches when the food was offered.

In short, whether or not we are physiologically hungry, when we are served food, we eat it — usually in its entirety. It may be hard to understand how a species that behaves in such a way could have survived over hundreds of thousands of years. The answer is that until recently humans did not live in an environment in which palatable food appeared out of nowhere. People had to expend effort to get food. That meant they did not become overweight.

But in the modern developed world, our food environment has been turned upside down. In our ancestors' world, food was scarce and required effort to obtain. Now it is plentiful, tasty, cheap, and accessible with minimal effort. Furthermore, we now have an enormous variety of choices, and variety encourages intake. And the menu includes "super foods," developed through millennia of cultural advances and modern food technology. Chocolate is a prime example: Nothing in nature provides the sweetness, fatty texture, appealing aroma, and calorie density that a piece of chocolate delivers.

It is an irony of eating that one has to expend energy, in foraging for food, in order to gain energy. Animals are designed to spend the least amount of energy possible to get the maximum energy from eating. They are optimal foragers, or energy misers. That holds for precultural humans as well.

There is no doubt that a food-regulation system exists in humans. It has been demonstrated in the laboratory, and it is revealed by the fact that, even with food readily available, many people keep their weight relatively constant for years without monitoring how much they weigh.

But in the developed world, food has become so palatable and so accessible, and requires so little energy to find and eat, that our natural regulation system is overwhelmed. Scientists have shown the same effect in rats when they are exposed in cages to abundant amounts of very appealing food.

Enter obesity. To our own peril, we can now succeed at what we evolved to do: Take in energy while expending virtually none.

In the last decade or so, a number of psychologists, nutritionists, and physiologists — including Kelly D. Brownell of Yale, James O. Hill of the University of Colorado, David A. Levitsky and Brian C. Wansink of Cornell, Marion Nestle of New York University, Barbara J. Rolls of Pennsylvania State University, and our group — have come to the conclusion that, in the developed world, increasing obesity can be attributed primarily to the environment in which we live and eat.

It is remarkable how even small environmental changes can lead to large increases in weight, as most changes occur over and over: every meal, every day, every year. An increase in food intake of simply one apple a day, for a year, with everything else constant, will lead to an increase in weight of about eight pounds a year. On the other hand, walking an extra block a day for a year, with everything else constant, will produce a weight loss of about five pounds a year.

In the developed world, food has become so palatable and so accessible, and requires so little energy to find and eat, that our natural regulation system is overwhelmed.

American adults gained about 1.4 pounds a year over the 10-year period from 1991 to 2000. That amounts to eating less than an extra quarter of an apple a day, holding other consumption and the level of energy expenditure constant. Thus the undeniable increase in obesity in the developed world could actually be eliminated by very small changes in habits or the environment. The so-called obesity epidemic need not be thought of as a deep sickness in modern society, but rather an undesirable trend that could be altered with modest changes, like replacing one sugared soft drink a day with a diet soda, or parking a block farther from the office.

American suburban life — with the car next to the door; remote-controlled appliances; the almost continuous availability of high-calorie, palatable snacks; the shift from daily food acquisition to a weekly trip to the supermarket; and ever-increasing portion sizes — encourages energy imbalance.

Does it have to be that way? The answer is clearly no. Perhaps the best illustration of that is another wealthy, developed country: France. The French are much less fat than Americans — according to recent figures from the World Health Organization and the U.S. Centers for Disease Control and Prevention, 8 percent of French adults qualify as obese, compared with about 31 percent of Americans.

Obviously that is not because French food is unappealing. Nor can it be because French food is lower in calories. In fact, the French diet has a higher percentage of total fat and saturated fat than the American diet. So how do the French do it? Our group, in collaboration with Claude Fischler, of France's National Center for Scientific Research, has found several interesting points.

One is that although the French spend more time eating each day than Americans do, they have fewer eating occasions. Almost all of their eating occurs at two meals — extended lunches and dinners.

In addition, the French pay more attention to the experience of eating — the taste and feel of the food in their mouths — than to the consequences of eating for their arteries and waistlines, and as a result feel less ambivalence about food than Americans do. The French eat more fat than we do, and they enjoy their food more.

But although eating fewer snacks helps to limit calorie intake, neither longer meals nor greater unadulterated pleasure in eating can explain why the French are thinner.

To understand the smaller French waistline, we have to appreciate a fundamental French-American difference. The French prize moderation in consumption, and quality in their diet. There are very few, if any, all-you-can-eat restaurants in France.

Americans put more emphasis on abundance and quantity; love and caring are expressed by offering a lot of food, as much as by providing high-quality food. Just consider our most famous holiday meal, Thanksgiving dinner — the success of which is gauged by how stuffed the guests are afterward.

It is also likely that the French expend more energy in an average day than do Americans. French people have fewer energy-saving gadgets, and most of them live in an environment where it is easier to shop for food and other goods on foot than by car, with many local bakers, butchers, and greengrocers easily accessible.

One reason for the lower French body weight that we have been investigating is portion size: The French serve less food at meals than Americans do, and hence they eat less. Even McDonald's in Paris serves less, with the same menu items containing fewer calories — for instance, there are fewer fries in a serving and fewer ounces in a large soda. The same is true at pizzerias, Chinese restaurants, and other fast-food emporiums in France.

The so-called obesity epidemic need not be thought of as a deep sickness in modern society, but rather an undesirable trend that could be altered with modest changes.

In addition, food in supermarkets that comes in individual portions, such as single-serving yogurt cups, has fewer calories. When we compared supermarkets, we found that the most common yogurt sizes in Paris and Philadelphia are 127 and 225 grams, respectively. Even French fruits are smaller, so when a French person and an American eat an apricot, the French person consumes fewer calories.

Many investigators have shown, in the laboratory or at eating establishments, that people eat less when they are offered smaller portions — even when they can have second helpings. That is a remarkable fact. If the French ate two yogurt containers, or two apricots, they would lose the advantage they gain from small portion sizes. What stops them from doing that?

Basically, consumption norms, ideas about what is a proper portion, are responsible. People commonly assume that a proper portion of something of at least modest size is one. It is OK to eat a very large sandwich — say, a 12-inch submarine — but odd to eat two 6-inch ones.

In a 2006 article with Gheorghe Doros in Psychological Science, we named that phenomenon "unit bias." We demonstrated it in a number of ways. For instance, we left pretzels in the lobby of an apartment house as free snacks for the residents. When we offered full pretzels (containing 300 calories each), people consumed 71 percent more calories than when we offered pretzels cut in half. Of course, people could take a second half-pretzel, but most didn't.

It is not only the food portion that is subject to unit bias, but also the serving instrument or receptacle: the size of the glass, plate, or serving spoon. At other times in the same lobby, we left out a big bowl of peanut M&M's with either a tablespoon or a quarter-cup scoop tethered to the bowl. A sign urged individuals to serve themselves as much as they wanted. With the scoop, four times as large as the tablespoon, people took 75 percent more candy. Of course, they could have dipped the tablespoon in repeatedly, but they tended not to do so.

One serving seems right. Unit bias therefore suggests that small reductions in portion size would reduce food intake.

There are two possible problems with using unit bias to help control weight gain. One is that people might compensate for smaller portions of one food by eating more of something else. However, the evidence from laboratory studies — particularly by Barbara Rolls and her colleagues — and from the lighter weight of the French suggests that people do not completely compensate for reduced portion size.

The second is that the metabolism of people who reduce their intake may become more efficient and slow down, so that their weight would not decrease as much as expected. Although increased efficiency at using energy has been shown with long-term and substantial reductions in food intake (as with many diets), there is no evidence that it occurs with very small changes in portion size.

We believe that the so-called obesity epidemic could be halted by minor changes in the eating environment. Reducing the portion sizes of food sold in supermarkets or served in restaurants by an imperceptible amount (say, a few percent) would be a good first step.

If food companies and restaurants chose to cut their portion sizes by a small amount and charged the same, they would make a little more money, people would lose weight, and we would have a slightly healthier population. That seems to us like an irresistible idea.

Paul Rozin is a professor of psychology, and Andrew B. Geier a Ph.D. candidate in psychology, at the University of Pennsylvania.

This essay originally appeared in The Chronicle of Higher Education (Chronicle Review, April 6, 2007).

Image credit: "Parisian Streets" by Katherine Shade