Monday 21 May 2018

Yanny or Laurel? Why people are hearing different names when they listen to the same viral audio clip

Mike Wright Sarah Knapton Mark Molloy

The question is seemingly a straightforward one: Do you hear Yanny or Laurel?

Yet answer for the millions of people listening to the short audio clip since it appeared on Twitter yesterday is proving as divisive as the infamous gold and white dress debate of 2014.

The seconds-long soundbite has provoked intense debate on social media as well as in offices and living rooms across continents, as people adamantly contest that their version is indeed what the synthetic voice is saying.

Academics have also offered a number of explanations for the baffling phenomenon. Theories range form the way our hearing changes we age, to people’s brains being primed by seeing the written word first, to accent variation between British and American listeners influencing what they pick up.

The debate started after the clip was first posted on social media site Reddit on Monday by user RolandCamry after he looked up the word “laurel” in the online dictionary Vocabulary.com.

RolandCamry, who wished to remain anonymous, told the Telegraph that after hitting the website's audio pronunciation function to hear how the world should sound his sister told him she heard it as “Yanny”.

To settle the debate he posted the audio clip on a Reddit forum posing the simple question: “What do you guys hear?”

The comments soon blew-up with posters arguing about whether it was saying Yanny or Laurel. One commented: “I hear Laurel and everyone is a liar”.

The clip then went viral when it was tweeted on Tuesday by YouTuber Cloe Feldman and has since been listened to more than five million times.

The conundrum has also been vexing celebrities such as talk show host Ellen DeGeneres and American model Chrissy Teigen, who have both weighed in with their thoughts.

Age gap

One factor that could be affecting what people hear is age as our ears become less adept at picking up higher frequencies as we get older.

This means that younger people could be more attuned to hearing the higher pitch Yanny rather than the lower pitch Laurel.

Radio broadcaster Lauren Laverne tweeted that she was hearing ‘Laurel’ whereas her seven-year-old daughter was hearing ‘Yanny’.

Professor Charles Spence, professor of experimental psychology, at Oxford, said he could only hear 'laurel' and it could be because of  his age.

He said: "There are some effects where words alternate what you hear when you say them rapidly repeatedly….. but given this is only said time two, that doesn’t seem like it.

"There are plenty of auditory perceptual grouping effects where you get one interpretation or another and it flips back and forth but again that doesn’t quite sounds like it.

"If there was stuff going on at high frequency range maybe you would get young people hearing/and being influenced by that, but not oldies?"

Accent variation

Jane Setter, a professor of phonetics at the University of Reading, said people may also be hearing different names depending on where they live.

“There may well be differences between British and American listeners,” she told the Telegraph.

“The phenomenon seems to have started off with American listeners’ perceptions of a recording of “Laurel” generated by computer i.e. synthesized.

“Accent differences will predispose the brain to hearing certain patterns so it would be interesting to see if a Brit, Aussie, American pattern emerges – but I’m not sure who’s collecting that data, and it will be difficult to do so now with any research rigour.”

Another factor that could influence what name listeners hear according to Prof. Setter was seeing the written word first in the tweet.

As English speakers read from left to right this could prime the brain to expect to hear “Yanny” for some people.

Sound quality

Alex Holcombe, a cognitive psychologist at the University of Sydney, has also explained that the audio quality of the clip could be an influence as our brains effectively fill in the blanks when presented with low sound quality. 

He said: “Because our brains are almost every day trying to understand what was said under less than ideal (noisy) conditions, it is in the habit of making strong guesses from ambiguous stimuli,” he explained.

“If the auditory signal is somewhere between the prototypical way of saying ‘laurel’ and ‘yanny’, then the brain may tend to force it towards one or the other, as it does with the phonemic restoration effect.”

Telegraph.co.uk

Entertainment Newsletter

Going out? Staying in? From great gigs to film reviews and listings, entertainment has you covered.

Editors Choice

Also in Entertainment