CAMBRIDGE, MASSACHUSETTS—A hush fell over the audience last night as Joseph Keller, a soft-spoken, white-haired mathematician, approached the podium. He held in his hand a red felt fez hat. "I used to jog around the Stanford campus and saw many young ladies running," Keller said. "Their ponytails swayed side to side like this," demonstrating with the fez's golden tassels, "even though the head was only going up and down. Why did the ponytail go side to side?"
There can be only one scientific event where questions like these are taken seriously: the Ig  Nobel Prize C eremony . Each year, just weeks before the actual Nobel prizes are announced, the Ig Nobel prizes are awarded for "research that makes people laugh, and then think." The raucous ceremony is presided over by Marc Abrahams, the editor of the Annals of Improbable Research .
Keller's deceptively simple question led him down a rabbit hole of mathematical modeling and mechanical experiments. A ponytail may look like a single well-behaved object, but it contains hundreds of thousands of individual fibers, each exerting elastic forces on the others. This mathematical modeling nightmare is known as a "many-body problem." But with help from a team of physicists, Keller discovered a set of surprisingly simple equations  that govern the shape and behavior of the swaying hairdo he observed on the Stanford University campus, where he is a professor of applied mathematics.
"It turns out that it could go up and down," Keller explained. "But that's unstable if the jogging frequency is twice the pendulum frequency of the ponytail," which is usually the case for humans. Because runners all tend to pound the pavement at roughly the same frequency and their hair is roughly the same length, all ponytails sway from side to side. This insight won Keller and his team the coveted 2012 Ig Nobel prize for physics.
This year's ceremony was marked by even more surprising twists than usual . Keller was awarded a second Ig Nobel prize last night—one that was, it turns out, long overdue. He had been a co-author on a study of the physics of biscuit-dunking  that won a 1999 Ig Nobel prize and for which he was not credited.
The other Ig Nobel prizes went to studies that had surprising twists of their own. Research led by Craig Bennett of the University of California, Santa Barbara, grabbed the neuroscience prize. Bennett's team wanted to demonstrate a problem with functional brain imaging: Too many researchers use the tool to go on "fishing expeditions," applying statistical techniques to the data that give misleading results, they noted. In the ultimate illustration of such bad fishing expeditions, the team applied those same techniques to "prove" that a dead salmon has brain activity .
Scientists honored with Ig Nobels did their best to remain dignified on the stage, although it couldn't have been easy. A giant inflatable clownfish drone hovered over the stage, colliding with walls and people alike. Meanwhile, a pair of Japanese researchers deployed their SpeechJammer , which thwarts people from speaking by projecting their voice back at them with a tiny delay; the invention won the researchers this year's Ig Nobel prize in acoustics.
But what would have been the most spectacular display of science was called off at the last second. The Ig Nobel prize for medicine went to a team led by Emmanuel Ben-Soussan, a medical researcher at the Clinique de l'Alma in Paris, for developing a technique that prevents the rare but dangerous problem of gas explosions during colonoscopy . (Yes, intestines really can explode .) A human subject was brought on stage for a demonstration. The doctor got as far as putting on a rubber glove before the demo was canceled for fear of offending the audience.