Â鶹ӰÊÓ

TORONTO -- A resurfaced audio clip that can be heard as two different words is dividing social media users once again.

In the video clip, , a voice can be heard either saying "green needle" or "brainstorm." The video ask viewers to listen to the same sound bite twice, once while looking at a caption reading "green needle" and again while looking at one reading "brainstorm."

According to the video’s caption, "you will only hear the word you're reading."

Jessamyn Schertz, an assistant professor at the University of Toronto's Department of Linguistics, told CTVNews.ca in an email that what an individual hears can change depending on which word is being read on the screen and therefore being thought about.

"Each time a word is spoken, the physical properties of the sound that is produced are very different --this is due to physiological differences between different talkers, or different speech styles, and different listening environments," Schertz said Thursday.

"All of this variation means that there is not a unique 'signature' that defines a given word, but that there's a wide range of possible sounds matching a word," she added.

Explaining the auditory illusion, Schertz said one’s mind can make a quick switch between the words by thinking about the second phrase while listening to the audio file, causing the ears and brain to latch on to the second perceived acoustic pattern despite previously hearing the first word.

"Humans are so used to listening to speech that it seems easy and effortless, but actually, reconstructing someone's message from the sounds we hear is an extremely complex task," Schertz said.

"Just by looking at a word -- or even by closing your eyes and imagining a word -- that influences the way we hear the exact same sound."

The TikTok clip was posted earlier this week and has since been liked nearly 185,000 times. It is also being shared widely across Twitter.

However, this is not the first time the ear-teaser has circled the internet.

The audio clip was originally posted in May 2018 by a user called 'squidjeep.' The video shows a toy that when activated says either "green needle" or "brainstorm." However, it was later revealed that the clip was actually saying "brainstorm."

The clip is similar to one shared just a few days earlier that year, debating the words "Laurel" and "Yanny."

To understand speech, Schertz said one’s brain uses a combination of the physical sounds or "bottom-up" information, and expectations based on real-world knowledge or "top-down" information.

"If I'm in an anatomy class and I hear something ambiguous between 'green' and 'brain,' I'm probably going to guess that it's 'brain.' Real-world knowledge that guides our choices can be knowledge about the context [or] expected meaning, but it can also be our knowledge or experience with the way a particular speaker says things," Schertz said.

According to Schertz, research shows that people adjust how much they rely on those different sources of information based on how useful they are.

"When the bottom-up, sound-based information is not as reliable, maybe because the recording quality wasn't great or because we're in a noisy environment, humans actually start to put more weight on that top-down information, or our expectations," she explained.

"This demo is a great example of how powerful those expectations can be."

Interestingly, some people listening to the audio may insist they are hearing "green needle" despite being wrong. Schertz said this is because the range of possible sounds can overlap in one's brain, falling into the possible domains of both words.

However, Schertz acknowledged that people's expectations when it comes to physical sounds are limited.

"Note that our expectations aren’t completely in charge… You're never going to hear 'hippopotamus' no matter how much you try to think about it."