Eat, Drink, and be Monitored

Ponca City, We Love You writes "A new restaurant has opened at Wageningen University in the Netherlands, fitted with a control center and two dozen hidden cameras devoted to exploring the question of what makes people eat and drink the way they do. Over the next 10 years, a team of more than 20 scientists will use the research facility to watch how people walk through the restaurant, what food catches their eye, whether they always sit at the same table and how much food they throw away. Researchers will examine environmental influences on eating behavior by making small changes in the color of the lights, in accompanying sounds, in the scents or the furniture. "We want to find out what influences people: colors, taste, personnel," said one researcher. "This restaurant is a playground of possibilities. We can ask the staff to be less friendly and visible or the reverse." University staff who want to eat at the new restaurant will have to sign a consent form agreeing to be watched."

Re:Science?

[gotzero]

Most university affiliated learning centers/experiments like this usually have heavily subsidized costs of goods. At my university, there was a restaurant that the hotel and restaurant management students ran. It was the best food and atmosphere for miles around, almost free, and completely not advertised. A lot of people did not know about it, and I ate there as often as I could get a reservation. I would have been happy to sign a waiver that my behavior could be watched there. I am sure a lot of my behavior was watched anyway, b/c the student staff were way too friendly to not be answering immediately to someone... Researching human behavior is a tough thing to tackle. Obviously the best data would come from blind or even double blind tests, but any kind of waiver or admission of experiment will obviously influence the subjects. The fact that people act differently when watched is enough on its own to show that there is still a ton to be learned on the subject.

Re:Scum

[Anonymous Coward]

But isn't the "thousands of external stimuli" a reason why this type of study is so problematic in the first place? This study assumes people are non-individuals with no particular history from the start. Unless you have a case history of every single person who is being studied the ability to tell why they might make certain choices is impossible. Studies like this look at inputs and outputs and don't concern themselves with the personal psychology of the subject. If you create a society governed by taste-tests then you create a society of non-individuals.

Re:Scum

[Albanach]

Nonsense, they don't care about the individual, they care about statistically significant shifts in the group.

Think about it, if burgundy carpets makes ten percent of the customers purchase a more expensive salad, and with no identifiable negatives, then it makes sense to install burgundy carpets if they want to shift more of these salads. It doesn't matter that it has no effect on 90% of the customers - indeed they would be well aware that it has no effect on them.

Stores run promotions all the time that are aimed at shifting a tiny proportion of their customers to a more expensive product. It doesn't work for the majority, but increasing your profit per customer for even a small proportion makes sense if you can do it without detriment to the majority of your customer base.

Similarly, while you may guess at why people made a choice, there's no need to know exactly why, just that you can record a statistically significant shift in their patters when you change one stimuli.

Re:on privacy

[Anonymous Coward]

Yes there is a sign on the door...

Having the experience of living in the Netherlands as foreigner for the last 5 years or so, I can tell you that the tolerance and support for surveillance in this country is scary. Very scary.

It comes AFAIK from a tradition of religious control, where people's lives were very closely followed by the religious. Go to any village in this country, and you will only see houses with huge windows and without curtains.

Nowadays everybody supports more cameras in the street.

This country has a culture of peer monitoring of behaviour, and peer "active" enforcing of acceptable behaviour (normally through the waving of a censoring finger, while preaching). Everybody watches everybody, and everyone will point in a censor-like way, to anyone not acting normally. This is no joke. The saying is "act normally" ("Doet normaal!"), for anyone doing anything nor conforming to "morally & socially" approved behaviour.

I could go on... but I better not.

Re:if you know

[CapsaicinBoy]

Disclaimer: IAAFS

Without meaning to be rude, you are flat out wrong. It just so happens that I study ingestion behavior for a living. My work is more related to the genetics of eating behavior and food choice, so this facility is less directly useful to me personally, but it absolutely will move the field forward. Unlike armchair quarterbacks that take cheapshots on the intarweb, every practicing scientist recognizes the inherent tradeoffs between experimental control and generalizability.

First, before the Correlation !=Causation weenies get their panties in a bunch, I'm happy for you that you passed stats 101, but you need to understand that RCTs are not the only way to do science. Yes, randomization is really nice for making claims about causation, but at least in humans, I can't assign you a specific gene (TAS2R38) or personality trait (novelty seeking). Yet we can still use the scientific method to make predictions based on theory and test those predictions.

Second, much of this work is done today using self report. Certainly, observation can induce bias, but so can self-report. When separate methods, with separate flaws confirm the same findings, science moves forward.

Finally, your comment about blinds, controls and isolation of variables is totally ignorant. The ability to manipulate this artificial restaurant in ways you could never manipulate a real restaurant is *exactly* what provides those controls.

Here is an example. Imagine I have a theory how socialization influences the time people spent at the table and the amount they consume. In this restaurant, I can manipulate the table size (2 vs. 4 chairs), social attachment between people (sit with friends or random assignment) or gender (do women eat more or less when seated with random men, male friends, just women, etc) to test my theories.