"Move to chopsticks!" he exclaimed, making bites smaller and harder to take. If the chopsticks are a bit extreme, smaller plates and utensils might work the same way. Study after study shows that people eat more when they have more in front of them. It's one of our predictable irrationalities: We judge portions by how much is left rather than how full we feel. Smaller portions lead us to eat less, even if we can refill the plate.
Speaking of which, Ariely suggests placing the food "far away." In this case, serve from the kitchen rather than the table. If people have to get up to add another scoop of mashed potatoes, they're less likely to take their fifth serving than if they simply have to reach in front of them.
"Start with a soup course," he says. That is what economists refer to as a default: Rather than putting everything on the table for people to choose, you begin by making the choice for your guests. If the first course is relatively filling and relatively low in calories, everyone will eat less during the rest of the meal.
Indeed, it's not a bad idea to limit the total number of courses. Variety stimulates appetite. As evidence, Ariely brings up a study conducted on mice. A male mouse and a female mouse will soon tire of mating with each other. But put new partners into the cage, and it turns out they weren't tired at all. They were just bored. So, too, with food. "Imagine you only had one dish," he says. "How much could you eat?"
What you eat, of course, is also important. Studies show that people aren't very consistent in the amount of calories they eat each day, but they're very consistent in the volume of food they eat each day. Thanksgiving is an exception to that consistency, but probably not to the underlying rule. Satisfaction doesn't depend on caloric intake; low-calorie, high-fiber foods and foods high in water content are filling. Thus, the more broccoli rabe there is at the table, the better.
Some cool wikipedia articles: