Imagine you are alone on a desert island for one year and you can have water and one other food. Pick what food you think would be best for your health (never mind what food you would like).


Alfalfa sprouts

Hot dogs




Milk chocolate

Paul Rozin, a psychology professor at the University of Pennsylvania, asked people this same question and found that 42% of people chose bananas, 27% spinach, 12% corn, 7% alfalfa sprouts, 5% peaches, 4% hot dogs, and 3% milk chocolate. Only seven percent of people chose a food that could actually offer them enough calories and all the nutrients they needed for long term survival. No, not alfalfa sprouts (not nearly enough calories): hot dogs and milk chocolate. These two animal products (the milk in milk chocolate) provide protein and fat, two necessary nutrients that would be deficient in the other foods. Overall, hot dogs would provide all necessary nutrients, sufficient protein, and a more optimal amino acid balance, suggesting they would be best suited to help you survive for a year.

So why is this psychology professor asking people what type of food they’d want to have if they were stranded on a desert island? Rozin found that people’s beliefs about what makes up a healthy diet is heavily influenced by psychology. In this particular paper (Rozin, Ashmore, & Markwith, 1996), he and his colleagues researched whether people’s views about a healthy diet were biased by something termed “dose insensitivity.” Before I describe what that is, let’s try out a few more of the questions that he asked people.

1. Do you agree with the statement: “Although there are some exceptions, most foods are either good or bad for health”?

2. For each option below, which diet is healthier (both diets have same number of calories):

-A diet that is completely salt-free, or a diet in which you have a pinch of salt every day?

-A diet that is completely fat-free, or a diet in which you have a pinch of fat every day?

3. For each option below, which food has more calories:

1 ounce of chocolate or 5 ounces of bread

A teaspoon of corn oil or half a teaspoon of pure animal fat


So, do you think foods are generally either good or bad? 

40% of people in Rozin’s sample did. This tendency to dichotomize foods into either being good for one’s health of bad for one’s health seems to bring with it particular “good food” and “bad food” schemas. Good foods are thought to be inherently low in calories and complete in nutrients. Bad foods are thought to be low in nutrients, high in calories and probably filled with heart-unhealthy things. The problem with this? This categorical thinking could lead to avoiding foods that are “bad” in large doses but are essential in small amounts. Also, in another study Rozin found that American’s associate chocolate cake with guilt, whereas the French associate it with celebration. Demonizing foods could lead to guilt, obsession with eating (termed orthorexia), and unhealthy eating habits.

Too much salt is better than no salt at all

Does categorizing foods actually lead people to assume even small amounts of certain foods are unhealthy? 

Rozin’s results say that for some people, yes. Over a quarter of people (26%) believed a salt-free diet was better than a pinch of salt each day, and nearly a third (31%) believed that a fat-free diet was healthier. He calls this “dose insensitivity.” Salt and fat are essential parts of our diet, conferring health benefits in small amounts. Yet, a sizeable minority of people believed that a diet completely free of these was healthier than a diet with trace amounts.

In line with the good food/bad food categorization, Rozin also found that almost half believed an ounce of chocolate had more calories than five ounces of bread (45%) and half a teaspoon of corn oil had more calories than a teaspoon of animal fat (47%). This confounding of caloric density and caloric intake suggest that a large number of people do tend to assume “bad” foods always have more calories than “good” foods, regardless of dose size. If you think that “good” foods inherently have fewer calories, no matter how much you eat… you might just find yourself eating more calories than you realize.

I first came across Rozin’s findings when I read Michael Pollan’s In Defense of Food. I hadn’t expected to find social psychology research in a book on nutrition, but it made me realize that our cognitions and attitudes shape our beliefs about food as much as any “truth” about what is healthy. I definitely grew up with the beliefs that “protein is good” and “fat is bad.”  No way would I pick a hot dog to be the one thing I eat for a year – hot dogs are unhealthy! I know that we tend to categorize things, and with that comes a host of characteristics that we apply generally to anything that falls into that category (think stereotype). I’d just never thought about the fact that I might be stereotyping my food, assuming that if something can be unhealthy, it must always be unhealthy. This type of belief can create a very unhealthy view of food such associating delicious things like chocolate cake with guilt instead of celebration. Along these lines, there is now even a term (not yet recognized in the DSM) for people who have an obsession with healthy eating.

What did you think would be the healthiest food to eat for a year if you had only one type of food? Did you fall prey to “dose insensitivity”? What else do you think psychology can tell us about eating?

For more on the psychology of food, read about 5 Simple Tricks for Healthier Eating

The article:

Rozin P, Ashmore M, & Markwith M (1996). Lay American conceptions of nutrition: dose insensitivity, categorical thinking, contagion, and the monotonic mind. Health Psychology, 15 (6), 438-47 PMID: 8973924

You are reading

Between You and Me

Short on Sleep? Conflict Could be Bad for Your Health

Fighting with a partner increases inflammation when short on sleep.

Taking a Step Back Could Save Your Relationship

Research shows taking a distanced perspective during conflict has many benefits.

Hate Your Partner Some Days? Research Shows You're Not Alone

Research suggests we all hate our partners a little, even if we don't know it.