With due respect to Jon Stewart and everyone else mocking New York mayor Michael Bloomberg for wanting to ban soft drink cups larger than 16 ounces, you're wrong. Indisputably so, in one regard. Arguably so in another.

To understand why, we have to take a look at a basic assumption almost everyone makes.

Why do we eat and drink? Be-cause we are thirsty and hungry. It's obvious, isn't it? We drink when we're thirsty, eat when we're hungry, and stop when we these needs have been satisfied.

If that is true, Bloomberg's ban is both paternalistic and silly. Not only is this a politician treating citizens like children ("No more pop for you, Johnny! If you're still thirsty, drink some water.") he's doing it in a way that cannot possibly work. Ban KFC's popular 64-ounce "mega jug" and people will simply suck down four 16-ounce containers. And get just as fat. If that assumption is true.

But it's not.

A small mountain of research has shown that hunger and thirst have surprisingly little to do with the choices we constantly make about what to consume and how much. Far more influential are environmental cues, including seemingly trivial factors such as the size of a soft-drink cup.

In a study that bore a family re-semblance to an episode of Candid Camera, researchers asked university students to sit down at a big bowl of tomato soup and eat as much as they wanted.

For some students, that was all there was to the experiment. They consumed an average of nine ounces of soup.

But other students unknowingly sat down to a "bottomless bowl," which constantly refilled with soup via a hidden hose. They consumed an average of 15 ounces of soup.

The dramatic difference lay in the environmental cues. The first group of students watched the contents of the bowl decline, which told them how much they had consumed. This information was crucial to their decision to say "enough." The second group lacked that information and so they kept spooning it up.

The "bottomless bowl" experiment was conducted by Brian Wansink, a Cornell University marketing professor and leading researcher in the field. He calls this behaviour "mindless eating," but it's not literally mind-less. As always, the mind is hard at work. It's processing lots of information. It's making decisions. But most of that activity occurs outside conscious awareness, and while we can consciously re-view the conclusions the mind draws unconsciously, and modify or overturn them, most often we don't. So a more accurate term would be "unconscious eating."

The results of "unconscious eating" can be startling. In one experiment, Wansink invited people to a theatre to watch a movie. Naturally, popcorn was handed out. But some people were given medium-sized bags, while others got large bags.

After the movie, the researchers snatched back the bags and weighed what remained. The result? People given the large bags ate on average 53 per cent more.

You might think that's because this was tasty popcorn so people who had the opportunity to experience more pleasure took it. Or maybe the people who got the large bags were simply hungrier. Wansink's team thought of these objections, and others, and repeated the experiment in various forms, including one in which the popcorn was left out for five days to make it so stale it tasted like chips of Styrofoam. And they always got the same result. "People eat more when you give them a bigger container. Period. It doesn't matter whether the popcorn is fresh or 14 days old or whether they were hungry or full when they sat down for the movie," Wansink wrote.

And that's not only true of pre-pared foods. When people are asked to make spaghetti or other foods in a kitchen, the size of packages they get ingredients from makes a big difference to how much food they prepare. With spaghetti, Wansink wrote, "we found that people who were given the large package of pasta, sauce, and meat typically pre-pared 23 per cent more - around 150 extra calories - than those given the medium packages."

This is why portion sizes are a critical component of the obesity epidemic.

During the Second World War, Coca-Cola set up 64 bottling plants around the world to serve the U.S. military. We all know the bottles that came out of those plants. They are elegant and iconic. They are also small. They contained 6.4 ounces.

In recent decades, prosperity and abundance - plus massive U.S. government subsidies for agribusiness that make corn syrup ridiculously cheap - produced a spectacular increase in container sizes. It takes 10 classic Coke bottles to fill one of those KFC "mega jugs." And it isn't a Marine slogging through the jungles of the South Pacific who's drinking the mega jug. It's a grossly overweight, sedentary teenager, whose risk of diabetes and a shorter life goes up with every slurp.

In a sane world, corporations that sold such a grotesque product to adults would be considered shady and irresponsible; corporations that sold it to minors would be scorned and vilified like they were tobacco companies. These corporations would suffer, and change their ways. There would be no need to even consider government regulation.

But we don't live in that world. In this world, most people don't know, or refuse to accept, that something as simple as portion size can have a major impact on their behaviour and health. And corporations don't suffer for heaping crap on kids.

So we're left with government regulation.

Here, I think reasonable people can disagree. The sort of thing Bloomberg is proposing restricts the liberty of both sellers and consumers. That's a serious concern that should carry a lot of weight, pardon the awful pun. But, as Bloomberg notes, it's a very modest restriction. You wouldn't be able to buy a "mega jug," but you could buy four 16-ounce drinks. You could even pour those four 16-ounce drinks into your own 64-ounce cup and continue the slow degradation of your health, if you insist.

As I said, reasonable people can disagree. But, on balance, I think Michael Bloomberg is right.

All that being said, it's import-ant to note that Bloomberg has cited Wansink's research, and others, as the basis for his proposal, and Wansink responded that Bloomberg was misguided. Writing on The Atlantic web-site Thursday, Wansink noted that other research showed that, when people were overtly denied the portions they were used to, they tended to rebel, consuming more later, for example. Rather than banning giant cups, Wansink wants the mayor to talk with food companies to "discover new ways to better promote lower-calorie options."

I admire Wansink's work, but I think his conclusion is dubious. True, people will likely re-act badly to the disappearance of "mega jugs" at first. But portion norms are not fixed.

It wasn't so long ago, remember, that no one expected to be able to buy 64-ounce soft drinks. Or even conceived of such a thing. If "mega jugs" were to go the way of leaded gasoline, the banning of which was also fought and resented, they would some day be forgotten. Like leaded gasoline.

And don't forget that Bloomberg now has the undivided attention of the food corporations precisely because the world's most prominent mayor has threatened them with a big stick. They're scared. And far more likely make real, voluntary concessions than they would have if the mayor had gone to them, cap in hand, and begged.

Unlike Brian Wansink, Michael Bloomberg understands politics.