No one dances like you. Your mom knows it. Your friends know it. Even computers know it.
Now, before you get too excited, keep in mind that it doesn't mean you're a particularly good dancer. Only that the way you sway — and throw your arms in the air like you just don't care — is entirely unique.
In fact, according to a recent study, a computer can recognize your moves, regardless of the music genre, with uncanny accuracy. The thing is, the scientists, from the Centre for Interdisciplinary Music Research at the University of Jyväskylä in Finland, didn't set out to build dance-recognition software.
Instead, you might say, they moonwalked into the discovery.
For their study, the researchers asked 73 participants to dance to various musical genres — while a computer algorithm tried to determine what kind of music they were listening to. Was it reggae? Jazz? Metal?
It turned out the computer wasn't so good at distinguishing styles of music. It made the right guess only about one in three times. But it did surprise researchers with its accuracy in another regard: It knew exactly who was dancing 94 percent of the time.
"We actually weren't looking for this result, as we set out to study something completely different," lead study author Emily Carlson explains in a news release. "Our original idea was to see if we could use machine learning to identify which genre of music our participants were dancing to, based on their movements."
In their introduction to the paper, the authors suggest that one of music's most universal traits is its use of "a regular or isochronous beat that affords synchronisation."
"One of the most salient features is its tendency to make us move; the majority of people respond to hearing music with some kind of movement, from simply clapping to a beat to engaging in complex dance movement."
Of course, we don't dance the same way to every song. That would be, well embarrassing. As the authors note, dance moves are modified according to the volume of the bass drum.
Not only that but "the presence of kick drum and bass guitar uniquely related to the speed of head movement."
That's a very technical way of suggesting that we dance to the music we're given. The thing is, at the heart of the dance are movements — and whether they're tempered toe-taps or all out booty-shaking, they may be as identifiable specifically to you, as any other biometric data.
"It seems as though a person's dance movements are a kind of fingerprint," study co-author Pasi Saari notes in the release. "Each person has a unique movement signature that stays the same no matter what kind of music is playing."
Some kinds of music, however, made it a little harder for the computer algorithm to work out the dancer's identity. It was least accurate, for example, in the presence of heavy metal.
"There is a strong cultural association between metal and certain types of movement, like headbanging," Carlson explains. "It's probable that metal caused more dancers to move in similar ways, making it harder to tell them apart."
While the researchers don't imagine their work contributing to yet another way that computers can trample individual privacy — at least not in the increasingly scary realm of facial recognition — it does open the door for a refreshingly quirky scenario.
Just imagine arriving sleepy-eyed at your office cubicle, setting down your morning coffee and flicking on your computer. Then, instead of logging in with a few keystrokes, you headbang, do the mashed potato, and maybe throw in a side of Macarena.
Happy Monday, indeed.