On 14 May 2018, the Internet lost its collective mind over a four-second-long audio recording of a word that sounded to some like "laurel," but to others like "yanny." As it is wont to do, the Internet divided into tribes and factions immediately, promoting hashtags like "#TeamLaurel" and "#TeamYanny."
In this post, we will address both the origin of the recording, and the various scientific explanations for its conflicting interpretations.
Where Did the Audio Come From?
The Yanny/Laurel fight gained viral prominence when self-described YouTube influencer Cloe Feldman tweeted a video of an Instagram story, which itself was posted on Reddit. But the actual Yanny/Laurel debate originated in a literature class at Flowery Branch High School in Georgia, both New York Times and Wired reported. These reports confirm that the audio came from a Vocabulary.com recording of the word “laurel” — a wreath worn on the head, "usually a symbol of victory," the Times noted.
Wired tracked down the students who first debated what that recording was actually saying:
On May 11, Katie Hetzel, a freshman at Flowery Branch High School in Georgia, was studying for her world literature class, where "laurel" was one of her vocabulary words. She looked it up on Vocabulary.com, and played the audio. Instead of the word in front of her, she heard "yanny."
"I asked my friends in my class and we all heard mixed things," says Hetzel. She then posted the audio clip to her Instagram story. Soon, a senior at the same school, Fernando Castro, re-published the clip to his Instagram story as a poll. "She recorded it and put it on her story then I remade the video and posted it," says Castro. "Katie and I have been going back and forth and we both agree that we had equal credit on it."
Reddit user RolandCamry, a friend of Castro's, says he then took the video from Castro's Instagram and posted it to r/blackmagicfuckery. "I originally saw it on an Instagram story," says RolandCamry. "From there I put it on Reddit."
The High Frequency/Low Frequency Factor
From a physics standpoint, nearly all experts quoted on this subject are in agreement that the noises that produce the "yanny" portion of the audio rest in the higher frequency portion of the recording and the "laurel" noises are in the lower frequency portion. Video game developer Dylan Bennett posted a useful visualization to Twitter clearly demonstrating that if one were only listening to the high frequencies they would hear "yanny," while listening only to the low frequencies produces the intended word, "laurel."
Okay, you're not crazy. If you can hear high freqs, you probably hear "yanny", but you *might* hear "laurel". If you can't hear high freqs, you probably hear laurel. Here's what it sounds like without high/low freqs. RT so we can avoid the whole dress situation. #yanny #laurel ? pic.twitter.com/RN71WGyHwe
— Dylan Bennett (@MBoffin) May 16, 2018
The Times also created a tool to explore this factor.
What a person hears personally is a result of how our brains "pick up on and interpret these frequencies," Rory Turnbull, a professor of linguistics at the University of Hawaii, told Vox.com. Some experts, like Brad Story — a professor of speech, language, and hearing sciences at Arizona State University — argued in the same Vox explainer that humans generally focus on only three frequencies, and the lowest frequency is "absolutely essential" to create the L and R sounds in the word "laurel."
A third expert in that story — Benjamin Munson, a professor of speech, language, and hearing sciences at the University of Minnesota — broke it down in similar terms:
People might be able to focus on the higher frequencies — the Yannys among us — because they have really great headphones or very good hearing, Benjamin Munson, a professor of speech, language, and hearing sciences at the University of Minnesota, suggested.
“But for the rest of us with po’folk headphones and old-folk hearing, we just hear the lowest-frequency components,” he wrote in an email.
Poor Audio Quality Contributes to the Confusion
While Wired reported that the original recording was first made in 2007 by a professional opera singer, the audio presented in the viral posts is a recording of a recording, as reported by the Times:
[Roland Szabo, a Flowery Branch High School student who posted the audio to Reddit] said Wednesday that he was working on a school project and recorded the voice from a vocabulary website playing through the speakers on his computer.
This is likely the source of the noises some experts have labeled as "sounds humans generally [do not] make," and also, according to the Times, explains why it could be easier for a person's brain to be confused by the interplay of high and low frequencies:
Playing the “laurel” clip over speakers and re-recording it introduced noise and exaggerated the higher frequencies. Those higher frequencies may have led to confusion over whether the word was Laurel or Yanny.
Psychology Is Also a Factor
We spoke with Diana Deutsch, a professor of psychology at the University of California, San Diego who is an expert in "musical illusions and paradoxes." She told us that the whole Yanny/Laurel debate reminded her of a musical illusion she discovered called "phantom words," described on her website like this:
In a ‘phantom words’ demonstration, each track contains two words, or a single word composed of two syllables, and these are repeated over and over again. The same sequence is presented through both loudspeakers, but the tracks are offset in time so that when the first sound (word or syllable) is coming from the speaker on the left the second sound is coming from the speaker on the right; and vice versa.
Because the sounds coming from the two loudspeakers are mixed in the air before they reach our ears, we are given a palette of sounds from which to choose, and so can create in our minds many different combinations of sounds.
Such a jumble of noises mixed in the air is similar to the mixing and muddling of frequencies found in the viral audio. Deutsch said that the words people interpret are often related to what they are "primed" to hear based on what is on their minds, their linguistic background, and other factors:
I play my 'phantom words' to students in class, and as a group they hear many different words. Often the words they hear are related to what is on their minds. For example, for the first ‘phantom word’ that I post on my web page, when it’s close to exam time, a student might hear 'no brain'. Students whose first language is Spanish might well hear 'bueno'. It’s also very easy to 'prime' people by suggesting a word.
This later point has been echoed by other experts as well, who have noted that presenting the audio along with the words people are supposed to be hearing can also prime them to listen for one thing or another:
But all this confusion — those so-called "shenanigans" — forces our brains to fill in the blanks of how the clip should sound.
It’s possible that knowing there are two choices — "Laurel" and "Yanny" — preps us to hear one or the other distinctly. Or listeners could be affected by the language they speak, or the last thing they were listening to before they clicked on the meme.
These subtle, weird differences in how the human brain interprets sound is something scientists are constantly trying to better understand, Turnbull said. But for what it’s worth, he added, "this is a really cool auditory illusion."
The bottom line? The original audio is indeed of the word "laurel," but there are plenty of scientifically approved excuses out there for people who hear "yanny." And some lucky individuals, of course, can hear them both.