NEW HAVEN, Conn. — Each year, Halloween is a massive operation at Dean Karlan's house, drawing in members of his family, particularly his 13-year-old daughter, Maya.
Maya's enthusiasm for Halloween knows few bounds. She makes her own costumes: a castle with a working drawbridge, a full-color traffic light with a flashlight inside to switch signals. She's been known to spend the whole year thinking about her next get-up.
"Normally I plan my Halloween costumes the day after Halloween," Maya said. "I get very excited."
FOR THE RECORD:
Halloween research: An article in the Oct. 31 Section A about researchers who use Halloween trick-or-treaters as test subjects for behavioral science experiments said that Stonehill College is in Connecticut. It is in Massachusetts.
But this year, the bubbly eighth-grader is giving up her trick-or-treat time — and her idea of dressing as a toaster — to help out at home, where the real action is.
At Maya's house, the children marching inexorably through the darkness toward the promised candy will get their treats. But as they climb the creaking stairs of the Karlans' porch, they'll also do something else: become test subjects in a one-night science experiment.
No need to be spooked. Karlan, a Yale University behavioral economist, just taps into a rare study population with a guaranteed turnout: trick-or-treaters.
For seven years, hundreds of children have been lining up in front of his home to answer questions from clipboard-carrying college students, and then choose their candy. Though they may not realize it, the answers they give and the sweets they pick offer the scientists fresh insight into theories about children's thinking, development, even their politics.
Ninth-grader Cole Benoit lives down the street. He's been subject to three rounds of experimentation over the years, and thinks it's fun.
"It's kind of like a game," he said.
Most studies about Halloween focus on some aspect of the holiday — mostly on common fears, such as the rate of pedestrian deaths that day or the perceived dangers of candy.
But very few actually use Halloween as a platform to study economics and behavior, even though it's a unique opportunity to do so, said Dan Ariely, a behavioral economist at Duke University who has done some trick-or-testing of his own.
"The benefit of Halloween was that the kids came to us rather than we having to go to them, and we were doing it in a natural environment that fits in their lifestyle and mode of thinking," said Ariely, who published a 2007 research paper that used trick-or-treaters to study the irrational decisions people make when it comes to getting "free" goods.
"It's a great opportunity for data. The only thing it requires is that you will have a house where the kids come to visit," he said, "and the adults are not too suspicious."
At the Karlans' home, each year is different. Last year's study found that 38% of kids 9 and older who saw a poster of First Lady Michelle Obama chose fruit instead of candy — twice as many as those who made that choice after seeing Ann Romney, wife of 2012 Republican presidential candidate Mitt Romney. The study indicated that the first lady's "Let's Move" campaign, promoting healthy living for children, appeared to be reaching its target audience.
Earlier years have been more abstract. A 2008 experiment, which asked kids to choose between either transparent or opaque bags of candy, found that children with more generic costumes preferred the see-through bags — that is, they avoided ambiguity.
This year, the study aims to test whether some children are inherently planners — whether they planned their costume ahead of time or procrastinated until the last minute, and whether they have a plan for how they will eat their candy. They'll weigh their answers against whether kids choose fruit or give into the easy temptation of candy.