Many people eyeball how much they think they will eat and then pretty much keep eating until they have a visual cue – such as a clean plate or bowl – that it is time to stop. What happens if the plate is never clean, and the bowl is never empty?
We devised an experiment to find out.
Although you would never find the construction plans for bottomless soup bowls in the back of Popular Mechanics, here is how they are made. You take a nice four person restaurant table, check that the owner is not around, and you drill a big one-inch hole right through the table where a waiter would ordinarily set a soup bowl. (A better option is to buy the table and then drill the hole.)
You then drill another hole in the bottom of the soup bowl so that you can attach rubber tubing into the bottom of the bowl, poke the other end through the hole in the table, duct tape it to the underside of the table, and run it over to a six quart pot of hot soup. By adjusting the height of the pot of soup, people will be able to eat out the soup bowl all day and it will keep refilling itself. It will not refill itself back to the top, but it will make people believe they are putting a dent in the soup even though the bowl never completely empties. This equalizing of the level of the soup occurs thanks to some help from atmospheric pressure which keeps the liquid in the 24 ounce bowl at the same as it keeps the liquid in the 6 quart soup vat. At any rate, people can pretty much eat out of it all day, and they would never reach the bottom because the fill level in both would drop at the same rate.
The table seated four people. Two had their refillable bowls rigged up to separate six quart soup pots, and the other two had normal bowls which looked identical. Because of the clean-up involved, we only ran one four-person study each day (Monday-Thursday) for four weeks. This meant that after we were finished with practice trials, we needed to find 60-64 hungry lunch volunteers.
This might sound straightforward, but the practice trials were disasters. There were four kinks that needed to be ironed out:
1. The Tube. The first big kink was the fact that having a tube poking into the bowl lead lunchgoers to get suspicious about what was happening. With the help of mechanical engineering student, a brass bayonet mount was designed that was fited to the bottom of the bowl, so the refilling tube never poked up into the bowl and could not be detected by running your spoon over it.
2. The Bowl. The second kink was in keeping people from moving their bowl. Since these were good Midwesterners, this problem was solved by asking them to not touch their bowls so we could “Keep everything organized and consistent.” Whatever that meant.
3. The Story. The third kink was that people did not initially seem to act natural because they were trying to guess why they were getting a free lunch. This was solved by telling them that in exchange for lunch we would be asked a brief series of questions about their impression about the campus cafeteria at which they had been recruited. We also moved the study to the Spice Box, a dinner-only restaurant in which various recipes are often tested during other times of the day.
4. The Soup. The kink was that the bottomless bowls simply did not work. That is, until we discovered that the noodle soup we were originally using either clogged the tubes or caused the soup to gurgle and to frighten our otherwise cheery lunchgoers. We bought 144 gallons of Campbell’s Tomato soup, and started the study again.
Once we had the kinks worked out, we recruited 62 people for a soup lunch. We asked them to enjoy their lunch and said that afterward we would ask them some general questions about the cafeteria and the food it served. They were seated four to a table. Two people at each table had 24 ounce refillable soup bowls, and the other two had normal 24 ounce soup bowls which were filled all the way to the top of the bowl.
You might think that you were asked to join three semi-strangers at a table for lunch, there might be some uncomfortable moments. Not so, with college kids. All we had to do was to get them started by asking them what their plans were for the summer and the conversation flowed as fast as the soup.
After twenty minutes, we stopped the study and gave these lunch-goers a survey that asked them to estimate how many ounces of soup they had eaten, how many calories this came to, and how full they were on 9-point scales (1=Not very hungry; 9=very hungry). At the same time, the soup was drained from the bowls and the tubes and the pots and weighed to figure out exactly how much each one had slurped.
Out of 62 people we recruited, only two discovered what was occurring. One had bent under the table to retrieve a dropped napkin, and quickly pointed out the Borg-like tubing under the table to the rest of his lunch companions. The second person had a much more dramatic experience. Forgetting for a moment that he was not at a medieval banquet, this man tried to pick up his bowl and drink out of it like a soup-hungry Viking. As he lifted the bowl, it made a loud gurgle and the red, tomato soup filled tube slithered up through the table like a snake. This made the woman shriek and the man across from him tip over his chair as part of a knee-jerk escape. All the these eight people were dropped from the study. None of the other 54 people suspected a thing.
People eating out of the normal soup bowls ate about 9 ounces of soup. This is just about the size of a non-diluted Campbell’s soup can. They thought they had eaten about 200 calories worth of soup. People eating out of the bottomless soup bowls, ate and ate and ate. Many were still eating when we stopped them after twenty minutes. The typical person ate closer to 15 ounces, but others ate more than a quart – more than a quart. When one of these people was asked to comment on the soup, his reply was, “Its pretty good, and its pretty filling.” Indeed it is, if you eat three times as much as the guy sitting next to you.[ii]
Surely the average participants realized that they ate more from the refillable bowl. Absolutely not. With the exception of a couple of people like Mr. Quart Man, people did not comment about feeling full. Even though they ate 73% more, they rated themselves the same as the others – after all, they only had about half a bowl of soup. Or so they thought. When asked how many calories of soup they ate, the 127 calories they estimated was the same as it was for those people eating from the normal bowls. In reality, they ate had eaten an average of 268 calories. This was 123 calories more than their friends with the normal bowls.
When we serve ourselves, it is important to remember that we almost always eat everything we dish out. The key to eating less is to dish out less to begin with because there is a good chance we decide we do not want seconds. If something is on our plate, there is a tremendous pressure for most of us to finish it. As a result, the best defense is not to put so much on our plate to begin with.
Some people believe that they are pretty good at regulating how much they eat. They claim that they eat when they are hungry and they stop when they are full. They are probably wrong. They might have some standard eating habits that make them think they can regulate what they eat. In reality, however, their standards probably are not very accurate. Their eyes will fool their stomach almost all the time if given the chance.
What would happen if the supply of food we were given never ended? Would we even know when to stop? This is partly what makes Thanksgiving dinners, buffets, and Dim-sum restaurants such diet dangers. Regardless of how much we eat, we never seem to put much a dent in how much food there is. But, if we carefully target how much we want to eat before we started, overeating would be less of a danger.
We devised an experiment to find out.
Although you would never find the construction plans for bottomless soup bowls in the back of Popular Mechanics, here is how they are made. You take a nice four person restaurant table, check that the owner is not around, and you drill a big one-inch hole right through the table where a waiter would ordinarily set a soup bowl. (A better option is to buy the table and then drill the hole.)
You then drill another hole in the bottom of the soup bowl so that you can attach rubber tubing into the bottom of the bowl, poke the other end through the hole in the table, duct tape it to the underside of the table, and run it over to a six quart pot of hot soup. By adjusting the height of the pot of soup, people will be able to eat out the soup bowl all day and it will keep refilling itself. It will not refill itself back to the top, but it will make people believe they are putting a dent in the soup even though the bowl never completely empties. This equalizing of the level of the soup occurs thanks to some help from atmospheric pressure which keeps the liquid in the 24 ounce bowl at the same as it keeps the liquid in the 6 quart soup vat. At any rate, people can pretty much eat out of it all day, and they would never reach the bottom because the fill level in both would drop at the same rate.
The table seated four people. Two had their refillable bowls rigged up to separate six quart soup pots, and the other two had normal bowls which looked identical. Because of the clean-up involved, we only ran one four-person study each day (Monday-Thursday) for four weeks. This meant that after we were finished with practice trials, we needed to find 60-64 hungry lunch volunteers.
This might sound straightforward, but the practice trials were disasters. There were four kinks that needed to be ironed out:
1. The Tube. The first big kink was the fact that having a tube poking into the bowl lead lunchgoers to get suspicious about what was happening. With the help of mechanical engineering student, a brass bayonet mount was designed that was fited to the bottom of the bowl, so the refilling tube never poked up into the bowl and could not be detected by running your spoon over it.
2. The Bowl. The second kink was in keeping people from moving their bowl. Since these were good Midwesterners, this problem was solved by asking them to not touch their bowls so we could “Keep everything organized and consistent.” Whatever that meant.
3. The Story. The third kink was that people did not initially seem to act natural because they were trying to guess why they were getting a free lunch. This was solved by telling them that in exchange for lunch we would be asked a brief series of questions about their impression about the campus cafeteria at which they had been recruited. We also moved the study to the Spice Box, a dinner-only restaurant in which various recipes are often tested during other times of the day.
4. The Soup. The kink was that the bottomless bowls simply did not work. That is, until we discovered that the noodle soup we were originally using either clogged the tubes or caused the soup to gurgle and to frighten our otherwise cheery lunchgoers. We bought 144 gallons of Campbell’s Tomato soup, and started the study again.
Once we had the kinks worked out, we recruited 62 people for a soup lunch. We asked them to enjoy their lunch and said that afterward we would ask them some general questions about the cafeteria and the food it served. They were seated four to a table. Two people at each table had 24 ounce refillable soup bowls, and the other two had normal 24 ounce soup bowls which were filled all the way to the top of the bowl.
You might think that you were asked to join three semi-strangers at a table for lunch, there might be some uncomfortable moments. Not so, with college kids. All we had to do was to get them started by asking them what their plans were for the summer and the conversation flowed as fast as the soup.
After twenty minutes, we stopped the study and gave these lunch-goers a survey that asked them to estimate how many ounces of soup they had eaten, how many calories this came to, and how full they were on 9-point scales (1=Not very hungry; 9=very hungry). At the same time, the soup was drained from the bowls and the tubes and the pots and weighed to figure out exactly how much each one had slurped.
Out of 62 people we recruited, only two discovered what was occurring. One had bent under the table to retrieve a dropped napkin, and quickly pointed out the Borg-like tubing under the table to the rest of his lunch companions. The second person had a much more dramatic experience. Forgetting for a moment that he was not at a medieval banquet, this man tried to pick up his bowl and drink out of it like a soup-hungry Viking. As he lifted the bowl, it made a loud gurgle and the red, tomato soup filled tube slithered up through the table like a snake. This made the woman shriek and the man across from him tip over his chair as part of a knee-jerk escape. All the these eight people were dropped from the study. None of the other 54 people suspected a thing.
People eating out of the normal soup bowls ate about 9 ounces of soup. This is just about the size of a non-diluted Campbell’s soup can. They thought they had eaten about 200 calories worth of soup. People eating out of the bottomless soup bowls, ate and ate and ate. Many were still eating when we stopped them after twenty minutes. The typical person ate closer to 15 ounces, but others ate more than a quart – more than a quart. When one of these people was asked to comment on the soup, his reply was, “Its pretty good, and its pretty filling.” Indeed it is, if you eat three times as much as the guy sitting next to you.[ii]
Surely the average participants realized that they ate more from the refillable bowl. Absolutely not. With the exception of a couple of people like Mr. Quart Man, people did not comment about feeling full. Even though they ate 73% more, they rated themselves the same as the others – after all, they only had about half a bowl of soup. Or so they thought. When asked how many calories of soup they ate, the 127 calories they estimated was the same as it was for those people eating from the normal bowls. In reality, they ate had eaten an average of 268 calories. This was 123 calories more than their friends with the normal bowls.
When we serve ourselves, it is important to remember that we almost always eat everything we dish out. The key to eating less is to dish out less to begin with because there is a good chance we decide we do not want seconds. If something is on our plate, there is a tremendous pressure for most of us to finish it. As a result, the best defense is not to put so much on our plate to begin with.
Some people believe that they are pretty good at regulating how much they eat. They claim that they eat when they are hungry and they stop when they are full. They are probably wrong. They might have some standard eating habits that make them think they can regulate what they eat. In reality, however, their standards probably are not very accurate. Their eyes will fool their stomach almost all the time if given the chance.
What would happen if the supply of food we were given never ended? Would we even know when to stop? This is partly what makes Thanksgiving dinners, buffets, and Dim-sum restaurants such diet dangers. Regardless of how much we eat, we never seem to put much a dent in how much food there is. But, if we carefully target how much we want to eat before we started, overeating would be less of a danger.
RSS Feed
