We’re entering the science fair season, when schools teach children about science by making their parents conduct experiments. Across America, grown men and women will be watching crystals grow, grapes shrivel, pennies corrode, and scuzz form in petri dishes – except for the most ambitious parents, who, out of love for their children and a desire to see them advance to science fairs at the city, regional, national, global, and galactic levels, will build fully operational communications satellites and nuclear fusion reactors.

But it all starts with the question. Many students may worry that the best questions (like, “Why does a can of Diet Coke float in water while regular Coke sinks?”) have already been asked.

Balderdash. There are an infinite number of good questions. For example, there’s the one Vicki Silvers Gier, a psychologist at Central Missouri State University, asked a few years ago: What happens when you read material that has been incompetently highlighted with a yellow marker?

Scientists had already shown that students benefit from highlighting their own reading material. But Ms. Gier and a colleague were the first to show that students who read material that has been badly highlighted by someone else (with unimportant words marked and important words not marked) will have dramatically reduced comprehension – even if they’re warned in advance that the highlighting will be terrible. “You cannot ignore it,'” she says. “Even when you try to ignore it, it’s there.'”

Scientific conclusion: Stupid people make us dumber.

I met Ms. Gier at the recent meeting of the American Association for the Advancement of Science in St. Louis, where she attended a presentation by Marc Abrahams, founder and editor of the Annals of Improbable Research. Every year, Mr. Abrahams and his colleagues give out their version of the Nobel Prizes: the Ig Nobel Prizes. Ms. Gier is an Ig Nobel laureate from 2002, after her paper “The Effects of Pre-Existing Inappropriate Highlighting on Reading Comprehension” appeared in the journal Reading Research and Instruction.

Mr. Abrahams, who has a vast knowledge of improbable scientific literature, compares Ms. Gier’s work to that of two Cornell scientists who showed that one attribute of extreme incompetence is “that the person so afflicted is incapable of knowing that he is incompetent.” The study, titled “Unskilled and Unaware of It,” demonstrated that people who scored, on average, at the 12th percentile in tests of humor, grammar, and logic assessed themselves to be, on average, at the 62nd percentile. Incompetence at the extreme is a double-whammy, the authors declare: “Not only do these people reach erroneous conclusions and make unfortunate choices, but their incompetence robs them of the metacognitive ability to realize it.” (Which explains Washington in a nutshell.)

The Annals of Improbable Research is like an ongoing tribute to science fair projects with a demented streak. “It’s about stuff that first makes people laugh, and then makes them think. What they think is up to them,” Mr. Abrahams says.

He doesn’t ignore the humanities (Ig Nobels have gone to the Apostrophe Protection Society for its efforts to “defend the differences between plural and possessive,” and to the Lithuanian creator of the amusement park called Stalin World), but Mr. Abrahams is particularly gifted at discovering scientific marvels, such as:

* Two scientists who, since 1927, have studied the rate at which a glob of tar drips through a funnel. Answer: One drop every nine years.

* The inventor of an alarm clock that rolls away and hides, forcing you out of bed to turn it off.

* Three scientists who proved that Kansas is flatter than a pancake (under a powerful microscope, “a pancake appears more rugged than the Grand Canyon”).

* A student who investigated whether the five-second rule of eating food dropped on the floor has any validity. (It doesn’t. Food picks up floor germs in under five seconds.)

* Japanese scientists who trained pigeons to tell the difference between paintings by Monet and Picasso.

And, finally, perhaps pushing the limits of what we define as “science”:

* Troy Hurtubise, a Canadian who built ultra-thick body armor out of discarded sheet metal and then frolicked with a Kodiak bear. He also allowed a car to run him over at 40 mph, and let his friends push him over a cliff.

The point is, science doesn’t have to be boring. There are an infinite number of highly improbable scientific experiments just waiting for some child and/or parent to come along and ask a truly demented question.

But, kids, please don’t try anything crazy with body armor. It’s been done.