![]()
When Jim and Nora were in elementary school, both Greta and I worked challenging jobs, so we did whatever we could to save time. Instead of bringing lunches made by their parents, the kids bought hot meals at school. The school was proud of its cafeteria. Kids had credit accounts, which meant they didn’t have to carry lunch money to school (thus making them less of a target for bullies). The children were encouraged to make “healthy choices” instead of just getting a ladleful of mystery meat plopped on their trays.
After a few billing cycles, however, we noticed that Jim was spending more and more money. A complete lunch was supposed to cost about $3.50, but his bill was nearly $50 a week! We asked the cafeteria what he was buying and a printout was sent home. Here’s a typical day’s meal:
Chocolate milk (2)
Hamburger
French Fries (2)
Jello (3)
Needless to say, Greta and I soon resigned ourselves to making lunches for the kids. While “healthy choices” sounds appealing, if these were the choices our child was making, then we were going to choose for him.
Parents want their children to make good food choices — we can’t be there to decide for them all the time — but we also want them to eat well now, even when they don’t seem capable of making healthy decisions. And of course not all parents have the time or even the capacity to make good decisions for their kids, so for decades, the school lunch has been thought to be an important key to getting kids to eat better.
Unfortunately, many programs designed to get kids to eat well have failed. If children are rewarded for eating good foods, then what happens when they aren’t being rewarded? Requiring them to eat a certain amount fruits and vegetables can backfire: even if kids don’t mind the taste, too much of anything eventually ends up being unappealing. Comprehensive educational programs sometimes work, but require so much parent and teacher involvement that they aren’t always practical.
So Helen Hendy, Keith Williams, and Thomas Camise devised a simpler approach: give kids a choice of fruits and vegetables in addition to their preferred meats, carbs, and milk, and reward them for eating even a little bit. They trained experimenters to observe 346 first-, second-, and fourth-graders for 6 days to see what fruits and vegetables they ate when given a choice of one of two fruits and one of two vegetables. Just one-eighth of a cup was considered a serving.
Then they started their reward program. Each student got to wear a Penn-State lion-paw necklace when the experimenters were in their school. The experimenters watched the students eat in groups of 12, recording which children ate 1/8 cup of fruit or vegetables. Half the groups were rewarded for eating fruits by getting a hole punched in their lion paw. The other kids were rewarded for eating vegetables. At the end of each week, any child who had three hole-punches could cash them in for a small prize like a colorful pencil or a decal. Did the reward system work? Here are the results for first-graders:

The graph on the left shows fruit consumption for each group of four meals observed, first during the baseline, then for three rounds of rewards. As you can see, when fruit consumption was rewarded, the kids ate significantly more fruit. Fruit consumption increased for the first round when vegetable-eating was rewarded, then dropped off in subsequent weeks. Similarly, the graph on the right shows vegetable consumption. Once again, vegetable consumption went up significantly when eating vegetables was rewarded, but only increased temporarily when fruit-eating was rewarded.
So rewarding kids for eating a type of food does increase the amount of food they eat. An important aspect of this study is that even though the kids were given a choice each day, that doesn’t mean they got exactly what they wanted. They could only pick one of two fruits and one of two vegetables (and unlike Jim, they had to pick one of each; they couldn’t just feast on jello and French fries).
The researchers also asked the students to rate which foods they liked on three separate occasions: before the study, two weeks after the end of the reinforcement phase, and seven months after the end of the reinforcement phase. After two weeks, the kids had a significantly higher rating for the fruits they were served during the study compared to before the study began. Their ratings for vegetables also increased, but the difference was not significant. After seven months, however, the kids ratings for both fruits and vegetables had returned to pre-study levels.
Hendy’s team says this is an encouraging result. After all, in many other studies, children actually said they liked fruits and vegetables less after the study period. They suggest that extending the reinforcement period longer than they did in their study might lead to kids liking these healthy foods for a longer period of time.
They are backed up by other research suggesting that children need to be exposed to new foods 8 to 10 times before a preference for that food can be developed. This is something we saw in Jim and Nora when they were kids. If we introduced a new food, the first time they might not even touch it. After a couple of meals, they might try it but say they didn’t like it. After a longer period, they might even decide they liked it.
In any case, it’s easy to see how horrible the approach at Jim and Nora’s school cafeteria was. Yes, they gave the kids choices, but the choices were both healthy and unhealthy. Few kids are going to pick a vegetable when they could take an extra dessert instead.
HENDY, H., WILLIAMS, K., & CAMISE, T. (2005). “Kids Choice” School lunch program increases children’s fruit and vegetable acceptance Appetite, 45 (3), 250-263 DOI: 10.1016/j.appet.2005.07.006