The best path to a healthy weight may be a good night’s sleep.

For years researchers have known that adults who sleep less than five or six hours a night are at higher risk of being overweight. Among children, sleeping less than 10 hours a night is associated with weight gain.

Now a fascinating new study suggests that the link may be even more insidious than previously thought. Losing just a few hours of sleep a few nights in a row can lead to almost immediate weight gain.

Sleep researchers from the University of Colorado recruited 16 healthy men and women for a two-week experiment tracking sleep, metabolism and eating habits. Nothing was left to chance: The subjects stayed in a special room that allowed researchers to track their metabolism by measuring the amount of oxygen they used and carbon dioxide they produced. Every bite of food was recorded, and strict sleep schedules were imposed.

The goal was to determine how inadequate sleep over just one week — similar to what might occur when students cram for exams or when office workers stay up late to meet a looming deadline — affects a person’s weight, behavior and physiology.

During the first week of the study, half the people were allowed to sleep nine hours a night while the other half stayed up until about midnight and then could sleep up to five hours. Everyone was given unlimited access to food. In the second week, the nine-hour sleepers were then restricted to five hours of sleep a night, while the sleep-deprived participants were allowed an extra four hours.

Notably, the researchers found that staying up late and getting just five hours of sleep increased a person’s metabolism. Sleep-deprived participants actually burned an extra 111 calories a day, according to the findings published last week in The Proceedings of the National Academy of Sciences.

But even though we burn more calories when we stay awake, losing sleep is not a good way to lose weight. The light sleepers ended up eating far more than those who got nine hours of sleep, and by the end of the first week the sleep-deprived subjects had gained an average of about two pounds.

During the second week, members of the group that had originally slept nine hours also gained weight when they were restricted to just five hours. And the other group began to lose some (but not all) of the weight gained in that first sleep-deprived week.