Science Explains Why This Auditory Illusion Makes You Hear 'Yanny' or 'Laurel'
A few years ago, a photo of a dress—excuse me, The Dress—tore the Internet apart arguing over what color it was. Now, there's an audio version that's dividing us from our friends and families. This time, we're dealing with a short audio clip that some people hear as "yanny" while others hear as "laurel."
In the audio clip, originally posted on Reddit a few days ago, a robotic voice can be heard repeating a word several times. It then circulated on Twitter this week, where the debate rages on.
Here’s the thing: No one can agree on what the word actually is. Some hear “yanny” and others hear “laurel”—and some people say they can hear both, or that the sound of the word switches on them suddenly. More than 20,000 people have weighed in on Twitter and the post has now been viewed over 9 million times. But there’s no right answer according to social media—or actual explanations for why two people can listen to the same audio and hear vastly different things.
But seriously, what the heck is going on?
Even audiologists think this is cool. “It’s fascinating,” Alison M. Grimes, Au.D., director of audiology and newborn hearing at UCLA Health, tells SELF, explaining that there are a few reasons why you might be on Team Yanny or Team Laurel.
For starters, your expectations play a huge role here. “If we’re talking to someone and we expect them to say a particular word in a sentence because it’s logical, we’ll hear it with our brain even if the speaker says it unclearly or it’s noisy,” Grimes says. “We automatically fill in the missing sound to make sense.” So if all your friends say they hear one word, it's possible that you will hear that word because that's what you expect to hear.
If you have some degree of hearing loss, that can also make a difference, especially if you have a type of hearing loss related to frequency. Because the phrase "yanny" resonates at a higher frequency than "laurel," you might be more likely to hear "laurel" if you have some high-frequency hearing loss, for example.
That means that the way you’re listening to the sound matters. “If it’s high fidelity and it’s turned up at a higher volume, you’re going to perceive it differently than if you’re listening on your earphones or a cheap speaker,” Grimes says, because high fidelity systems allow you to hear frequencies more clearly.
This doesn’t mean that everyone who hears “yanny” over “laurel” has crappy hearing, Grimes says. Because there are so many factors that go into how you hear a specific sound, it's difficult to say exactly why one person might hear it one way while their friend hears it differently.
Some people say they can hear both words, and there’s a possible reason for that.
If you listen to a word over and over and you’re only given two choices, your brain immediately takes away all other choices and elevates the perception of only those two, Grimes explains. “If you know that your friend hears ‘yanny,’ you’re prepared to hear it," she says. "But you also help your brain think ‘How could I make ‘laurel’ out of that?,’ and all of the sudden you hear ‘laurel’ come through."
And, although The Dress had a definitive answer (it turned out it was blue and black IRL), Grimes likens this audio illusion to drawings where you could see two different things, like a vase or two faces, or a duck or a rabbit. “There isn’t a right answer because this is a digitized speech file,” she says. “When you digitize a sound file, you can manipulate it in any number of ways.” So with some clever tinkering, you can basically make yourself hear either option.
So the only right answer here is that there isn't one. The way you perceive the world isn't an exact representation of what it's actually like—your vision is not a camera and your hearing is not a recorder—because your brain factors in all sorts of other things that are individual to you, including your expectations and your previous experiences. So, the good (and infuriating) news is that whatever you, your friends, and your coworkers hear is correct.
Related:
