For consumers in California, New York City, Portland and Seattle, it might not come as a surprise that Starbucks's raspberry scone contains 500 calories or the foot-long meatball marinara sandwich at Subway has about 1,160. But because these few local and state governments have introduced mandatory menu labeling in chain restaurants, have people been cutting back on the calories in their orders?

Consumers across the country will start seeing these numbers on menu boards when labeling requirements roll out nationwide as part of the new health care reform law. Findings on this approach's effectiveness, however, have been decidedly mixed.

One small study conducted at an Asian-style fast food restaurant even tried to actively solicit customers to cut calories by asking if they would "like to downsize that" (rather than the now maligned "supersize that" offer) to cut 200 calories from their meals. About a third of consumers took the offer, but that rate did not improve when calorie counts were posted, reported New York University's Brian Elbel at the American Public Health Association meeting in 2010. This difference suggests that although some consumers might be willing to change their lunch routine with a prompt, calorie figures did not seem to tempt them away from "visceral urges" when they stepped up to place their order, Elbel explained.

A new yearlong study, published online January 14 in the American Journal of Preventive Medicine, found that consumers at one fast food chain in the Seattle area were unfazed by calorie counts listed with their favorite menu items.

In January 2009 King Country in Washington State, which includes Seattle and some of its suburbs, started requiring chains to make nutrition information available for all of its offerings, including visible calorie counts (which also needed to be on drive-through boards by August of that year). So researchers compared consumers' food choices at several locations of the Mexican-style restaurant called Taco Time before and after calorie numbers were posted—as well as with Taco Times that were outside of the regulated area.

After a year of looking at the labels Taco Time consumers were still not persuaded to ditch more of their beef Roma burritos (843 calories) in favor of regular chicken salads (196 calories)—overall order calorie totals (including sides and drinks) stayed relatively stable. Was this because health-conscious Seattle-area residents had already been able to make healthful choices via the chain's "healthy highlights" menu? Or is it simply foolish to think that a few posted digits might come between people and their nachos?

Scientific American spoke with one of the study authors, Eric Finkelstein, an associate professor at Duke University's Global Health Institute and in the school's Health Services Research Program in Singapore, about why the extra helping of information did not influence ordering behavior.

[An edited transcript of the interview follows.]

Even though customers at the King County restaurants had a full year to digest the calorie content of their orders, they didn't seem to start more healthful choices. Why do you think that is?
The fact that it was zero change was a little surprising. This information just doesn't change their behavior. I think fast-food customers who are busy and interested in having a nice meal are just not that interested in the calorie posting. I think super healthy people just don't go to Taco Time.

I think Taco Time also made it easier for consumers to identify which were the lower-calorie entrees by having the "healthy highlights" menu. If you are health conscious, and you can buy a Coke, a Diet Coke or water, do you really need to know the calorie counts? You already know which has more calories.

Does that imply that people who are making less healthful decisions are less aware of the relationship between poor diet and poor health?
I find it hard to believe that in this environment people don't know that eating fast food is not healthy. Most people seem to know if they're overweight, and that hasn't seemed to change their behaviors. Obese people certainly know that their weight is increasing their risk of poor health—they actually overestimate the mortality risks of obesity. I think information alone is not going to change people's behavior and get them to lose weight.

If you look back at tobacco, the truth is that most of the change in smoking was due to cigarette taxes—the information campaigns had only a marginal effect. I think prices have a more salient effect when it comes to these types of behaviors.

All of this has reminded me a little of the lessons we've learned from behavioral economics—that as consumers we don't always make the best, most rational decision in our day-to-day lives. Do you think similar dynamics are at work here?
It's possible, or it could be that consumers just like the taste of the food that they buy and listing calories just doesn't change their mind.

The simple story is: consumers knew what they like, you gave them information, and they still buy it. It's actually not an irrational decision—it's got all the ingredients that people crave: sugar, salt, fats. Once you buy it and you start eating it, it's quite difficult to stop.

If calorie information didn't change orders at the restaurants you studied, do you think there are other types of establishments, situations or even locales where the extra information would make more of a difference?
There may be places where you would see a modest effect. But I'm skeptical that labeling could get people to make significantly different choices.

It may encourage suppliers to reformulate in a way that would be more healthful—and in fact, you're seeing that in a few locations. Even Taco Time, which has a relatively healthy menu, is looking for a way to reformulate to make options healthier—one of the best moves they made financially was to include white meat in their chicken products, because that's what consumers wanted. But right now, on average, consumers are not demanding low-calorie food options. If consumers really wanted it, makers would respond.

So if calorie counts don't seem to nudge people to more healthful ordering, what interventions do you think would be more successful?
I'm an economist, so if you want to get people to make sustained changes in their behavior, you need to make sustained changes in the environment. But we've changed the world and made it harder to engage in healthy lifestyles.

Personally, I like being able to see the calories. But a question is, on a broader scale: Do we need the government to regulate, or could the market serve the information that consumers desire?

I would focus on kids and schools, and would ensure that kids are getting healthy foods in school, and are getting physical activity. I think focusing on adults is an uphill battle. I think having employers take a lead in encouraging healthy work sites—with government support—may have a change in getting some sustained behavior changes.

What incentive do you think would work best?
I'm not sure I'm in favor of taxing fast food, but I'm sure it would work.

Short of passing a new tax or making larger changes to the eating environment, is there information that might be more effective than just the calorie counts?
I think the "healthy highlights" are probably sufficient and may be better than including all of the nutrition information. Faced with a lot of information, some people get confused. But if you had one logo to show healthy options, in theory, that could be better than having all of the nutrition information. We should be doing more research around that for sure.