Dutch diners double as laboratory subjects

At the university cafeteria, women linger longer than men over their lunch decisions. Given a choice, they tend to opt for meat labeled “animal-friendly.” Men are likely to go for a new product.

Cameras are pointed at them. From a control room, monitors record the customers’ movements, hesitations, facial expressions, posture, weight -- even their eating habits.

Does it matter if the cheese slices are wrapped in plastic? If the bread is presented as a loaf or sliced? Whether the salad is on a red table or a blue one?

The $4.5-million Restaurant of the Future is run by scientists from Wageningen University and Research Center, working with Sodexo, an international catering firm, and the Noldus software company, to answer questions from the food industry and behaviorists.

“We think of ourselves as rational beings, always making the best choice,” says Rene Koster, director of the Restaurant of the Future Foundation. But that’s not true; 80% of our decisions are made subconsciously, he said, citing U.S. studies.

Research on consumer behavior has been around since marketing began. But with its spy machines, databases and battery of analysts, the Wageningen project, with 42 companies participating, is meant to take the study of eating to a much higher level.

Knowing how to subtly guide choices could have a huge commercial impact. About half of all food consumed in the United States is eaten outside the home. The figure is 68% in Japan.

Companies are interested, of course, but so are public facilities. A hospital in Utrecht has asked for a study of the effects of a better meal or a change of dining surroundings on the well-being of its patients. Schools want to know how to deal with teenagers who throw away home-prepared food and lunch on potato chips and Coke instead.

The cafeteria is organized in a series of islands, each with a different food type, in what Koster called a free-flow system rather than the traditional long line serving everything. People feel they have a wider selection, he said, and they tend to spend more money.

The checkout is self-service. Flush with the floor at the checkout is a scale that records the customer’s weight.

The lunchroom serves about 200 diners a day. Some 480 people registered for the project and gave their consent to be monitored. Visitors are not part of the experiment.

Discreet ceiling cameras can zoom in on a face or a plate or pull back to view a table or broad section of the lunchroom. They record not only what food customers selected but what they almost selected and how long they paused before deciding. Facial recognition software analyzes their level of enjoyment.

“The first weeks are not very nice, but you get used to it,” said diner Vida Mohammadani, a campus employee, adding that she no longer pays attention to the monitoring.

Volunteers are motivated to sign up for the project by the cheap prices -- the cafeteria is one of the few places in the country where you can eat a full meal for $6.30.

Researchers are just beginning to experiment after establishing the clientele’s baseline behavior since the cafeteria opened early this year. But they already are piecing together anecdotal evidence. Some examples:

For months, he said, customers bought milk from a vending machine. One day the label was changed to indicate the milk was organic -- prompting some people to comment that it tasted funny.

People eat more when food is served on a big plate, less on a small one.

In a study published in the September-October issue of the Journal of Nutrition Education and Behavior, Wageningen doctoral candidate Pascalle Weijzen reported that people who say they like healthy food often succumb to junk. Nearly 600 people were asked which snack they would pick: an apple, a banana, a candy bar or a molasses waffle. A week later, they were offered the choices -- and 27% who had claimed to prefer fruit had the candy.

“This is a laboratory. We control all the conditions,” Koster said. The prices, assortment, arrangement and presentation can be changed according to scientific need.

“But we still call it a restaurant,” he said. “If we used the word ‘laboratory,’ it might influence behavior.”