Watching people beat in an MRI


If you have trouble controlling the rolling trill that differentiates pero of perro As much as I do in Spanish, beatboxing is probably not the right hobby for you.

Previous research has shown that performers regularly interpret their vocal muscles in forms foreign to their native language. And according to the results presented yesterday at a joint meeting of the Acoustical Society of America and the Canadian Acoustical Association in Victoria, British Columbia, many artists also produce sounds. completely unknown to modern linguists. You can now see exactly what produces these extra-linguistic effects.

You look at the flying boxer's tongue, throbbing like an alien worm as he rides a wave of pressure from the vocal cords of the throat to an explosive release of the lips. Cognitive science researchers at the University of Southern California call this a "slippery labial stop without glottalic without a voice," but most of us recognize it as the interpreter's approximation of a fat beating. checkout. thud.

And there is more where it comes from. The team captured MRI videos of various artists playing 30 other effects, ranging from thrilling bass drums ("pronounced unpublished glottal velvet stops") to vivid click-thrusts ("lingeless vocal alveolar trills"). You can even compare them side by side on the group's website.

So why does an interdisciplinary team of linguists, engineers and computer scientists watch beatboxers beat on their backs in an MRI machine? This skill straddles the intersection between language and music, said Tim Greer, PhD student at USC, at the conference yesterday, and his group hoped the new art form could help to mark the limits of what humans can accomplish linguistically.

"[Beatboxers] let's try to create a new sound system, "says Shrikanth Narayanan, USC engineer and team leader. "We just wanted to see the similarities and the differences from a cognitive science point of view."

And this effort begins by cataloging the collective repertoire of various artists. Previous research had focused on an interpreter, but this time, the band was able to recruit five different beatboxers, including one of their own, undergraduate biomedical engineer, Nimisha Patil.

During a session, each beatboxer spent between 30 and 90 minutes playing all the sounds he knew while an MRI machine was recording head and mouth slices. Beating in his back was not too hard, said Patil. The biggest challenge was to resist the temptation to keep pace, which would distort the imagery. "It will not be your head, it will be the middle of your eye or something," she says.

After filming precise movements of the tongue and various other parts of the vocal system, researchers have attempted to categorize them using the traditional linguistic framework to break down sounds according to how the mouth produces them and where. A "Bilabial Fricative", for example, comes from the friction of the air that blows between two lips – like the v the sound or the b sound in English. Greer said in his presentation that some rhythms fit perfectly with similar descriptions, but that others, like the "inner click-click with the lipstick" – completely broke the linguistic mold. "We do not see this articulation in any language we know," he said.

Lip rolls, distant cousins ​​of the lip and raspberries, are among the most difficult effects to learn, says Patil, which makes perfect sense given Greer's findings. "If you think about the language you speak, you will never roll your lips in the middle of a sentence," she says. We learn a language from the sounds we find around us. So you would not expect these rare effects to be captured by osmosis – with the exception of perhaps the most devoted Yankees fans.

Although the group does not yet have many concrete linguistic advances, it dreams of great things. Narayanan then hopes to rely on the "vocabulary" that they have cataloged to see if the practice has patterns that look like grammatical rules, such as if some of the effects tend to follow others. He sees the beatbox as a complementary system, overlapping but independent of speech, which can expand the field of linguistics.

"We can go beyond language," he says. "Beatboxing is another platform."

Patil is also interested in the linguistic aspect of beatboxing, but she also has a more practical application: to learn – and teach – new beats faster. She says that most beatboxers come out largely by trial and error, imitating others and playing randomly. But the cutaway vision of this technology exposes the exact machines behind each time, eliminating the mystery and need for riddles.

"Just to see the different movements of the language," says Patil remembering having first seen MRI images of her own beatboxing session. "There is so much going on that I did not even know I was doing it."