HomeSkepticism

March, 2004
In 1891, Charpentier published the first systematic report of what has been called the Size-Weight or Charpentier's illusion. A classroom demonstration of this illusion may be constructed with a large container, such as an empty 9.5 L water container, 13 film cans, a Kg of lead shot and a scale.

The spout is removed from the water container to reduce weight (it should weigh about 180 to 200 g) and the film cans are filled with the lead shot to provide a series of stimuli to allow people to attempt to match the weight of the water container.

A good series of matching weights would be (in grams): 5 (trim the bottom off), 10, 15, 20, 25, 30, 35, 40, 50, 60, 70, 80, 100, 120, 140, 180 (or whatever the water container weighs), 220, and 260. Label the film cans.

To perform the demonstration, first keep the scale hidden. Line the film cans up in ascending order on a desk with the water container behind them. Instruct the students that they are to lift the cans and cannister and report which of the film cans appears to weigh the same as the large cannister.

I do this individually, with students coming into the classroom one at a time to eliminate an audience or conformity effect. After the data are collected (usually by the first student to make a judgment, to keep things honest) construct a frequency distribution on the board. Get them to calculate the mean, median and mode. For more advanced students, ask them to do a statistical test and write up the lab.

When the swiftest student asks, "How do we know those are the real weights?", whip out the scale and ask her to perform the measurement.

Charpentier's Illusion at The Amazing Meeting 2

Outside of the meeting room at TAM2, I set up a table with two objects on it. One was an empty 9.6 L water jug weighing 175 g, the other, a 7.5 mL lead sinker weighing 85 g. A sign in back of the objects asked that people judge which object weighed more. When they made their judgment, they recorded it and signed their name (optional).

Judgments were made by 155 people and 26 correctly identified the water jug to be heavier, while 129 incorrectly picked the lead sinker. The Binomial test showed that the group as a whole performed well below what would be expected by chance (p<.0000001) Among the people who were correct were Jerry Andrus and Claus Larsen.

When I announced the results at the Sunday paper sessions Randi , addressing the audience, said, "I was wrong." Then another attendee said, “But your instructions were to tell which one felt heavier.” Nope. I wrote the instructions and I ought to know. This person was showing a false memory syndrome as well as Charpentier’s size-weight illusion.

Editor's note

I was one of the few people who judged correctly. But I figured they were about the same weight, except the jug felt a little heavier. I did not judge the water jug to weigh twice as much as the lead sinker. This is a great test for anyone who claims they can trust their own perception.

They most probably can't.

PubMed abstract: Charpentier (1891) on the size-weight illusion.
PubMed, a service of the National Library of Medicine, includes over 14 million citations for biomedical articles back to the 1950's.

Jeff Corey
Psychology Department
C. W. Post College