This episode was written and produced by Carolyn McCulley.
What we hear is incredibly personal and we all hear things differently. Sometimes our ears can even play tricks on us. Sonic illusions put a spotlight on the unique function of our hearing and how our backgrounds and biology affect how we process sound. Psychologist Dr. Diana Deutsch and neuroscientist Dana Boebinger explain why our hearing is a unique sense and why sonic illusions can fool us.
MUSIC FEATURED IN THIS EPISODE
All Coming Together by Dexter Britain
Future Hit (Instrumental) by Louis II
Future Hit by Louis II
I'm Not Here (Instrumental) by Graphite Man
Petite Suite: I. En Bateau by Sunshine Recorder
Pensive Robot by Eric Kinny
From Scratch by Chad Lawson
Every Passing Second by Max LL
June 3rd (No oohs and ahhs) by Virgil Arles
Twenty Thousand Hertz is produced out of the studios of Defacto Sound, and hosted by Dallas Taylor.
Follow the show on Twitter & Facebook.
Become a monthly contributor at 20k.org/donate.
If you know what this week's mystery sound is, tell us at mystery.20k.org.
Consolidate your credit card debt today and get an additional interest rate discount at lightstream.com/20k.
Check out SONOS at sonos.com.
View Transcript ▶︎
[SFX: Virtual Barber Shop Plays]
You’re listening to Twenty Thousand Hertz. I’m Dallas Taylor.
[SFX: continue Virtual Barber Shop]
What you’re hearing right now is the virtual barbershop. It was made by QSound Labs and uploaded to Youtube back in 2007. It was the first time I remember being pretty blown away by a sound trick. By the way, before we go any further, this episode is all about sound tricks and sonic illusions. It’s one of those episodes that will be much more effective in headphones, so if you have those handy, pause the episode and put those on.
[SFX: Virtual Barber Shop Continues]
Our ears are amazing things. And we all hear things differently. And sometimes our ears can even play tricks on us.
The Virtual Barbershop is created by using a binaural recording technique. This simulates the way our ears perceive sound putting the microphones in the ear canal, and using the actual ear structure to help shape the recording. That’s why it can feel so immersive over, say, a normal stereo recording. But, it doesn’t work for everyone. Our hearing is an incredibly personal sense. We all hear things slightly differently, and sonic illusions can really put a spotlight on these differences… and some sonic illusions can really mess with your mind.
[SFX: Shepard Tone]
This is the sonic illusion known as the shepard tone, named after cognitive scientist Roger Shepard. It gives listeners the impression that sound is constantly going up or going down in pitch but never resolving. The Shepard Tone is intended to evoke anxiety or tension in the listener.
[SFX: Shepard Tone End]
[SFX: Dunkirk soundtrack]
What you’re hearing now was a shepard tone in the score to the movie Dunkirk. The director, Christopher Nolan, based the whole film on the concept of the shepard tone – score and script alike. He explains this in an interview for the UK TV channel Film4.
[Christopher Nolan] “I approach structure from a very mathematical and geometrical point of view. And so the structure I settled on is based on a musical structure called a Shepard Tone, which is a musical illusion whereby you can keep climbing up a scale. You are continually going up the scale, going up the scale, but you never seem to go out of, out of reach, if you like. And I wanted to try to apply that to screenwriting, to a narrative, and say ok so with this story, can you braid together the three storylines in such a way that you create the idea of a continuing rise in intensity, narratively.”
So how do you go about creating a shepard tone? It’s achieved by stacking several ascending notes on top of each other [SFX: Shepard Tone]. Each separated by an octave. While lower notes are fading in at different times, higher notes are fading out, and that’s what makes it seem like it never resolves. You can also apply these same concepts to a musical scale.
[SFX: Shepard Tone End]
[SFX: Mario 64 Endless Stairs]
Fans of Super Mario 64 will recognize this shepard tone in the game’s “endless stairs.”
[End Mario SFX]
But the Shepard Tone is just one of many sonic illusions. Others can show us just how subjective sound can be. This has led to some serious debate on the internet.
[SFX: Laurel/Yanny music mashup]
So, is it Laurel or Yanny?
I heard Laurel, but the Twenty Thousand Hertz team reported hearing both. So we asked two experts in psychology and auditory cognitive neuroscience—and even they don’t agree.
Diana: I hear Yanny.
That’s psychology professor Diana Deutsch, a pioneer in audio illusions, squaring off against cognitive neuroscientist Dana Boebinger.
[SFX: boxing match “ding ding” bell]
Dana: I heard Laurel, and I almost always hear Laurel. But I'm okay with that, because it actually is Laurel.
Dana: The recording is actually from vocabulary.com, an online dictionary, and it's the online audio pronunciation for the word, "Laurel," it's recorded by a voice artist, an opera singer who was hired to record a bunch of pronunciations for the website. My take as to why it's so ambiguous, and why so many people heard Yanny when that's not actually what this man was saying, is that the recording wasn't super high quality. It was made in a DIY recording booth where this man presumably recorded thousands of these words.
So what’s the science behind the Laurel-Yanny debate? Dana believes that it has to do with the way we perceive frequencies.
Dana: I'm a PhD student at Harvard and MIT studying auditory cognitive neuroscience. So I study how the brain understands sound. The voice is actually made up of tons and tons of frequencies stacked on top of each other. So Laurel and Yanny are actually kind of similar, and Laurel has some of the sounds that might have beefier lower frequencies [SFX: pitched in lower frequency] whereas Yanny might have more emphasis on the higher frequencies [SFX: pitched in higher frequency].
But there is some disagreement about whether the Laurel/Yanny illusion is in fact caused by frequencies. Perhaps it’s caused by something else entirely.
Diana: I'm not sure that I go along with the frequency thing.
Diana is a professor of psychology at the University of California, San Diego. She is internationally known for the audio illusions and paradoxes that she’s discovered.
Diana: I think it's much more likely to be due to the patterns of speech that we hear in our particular language or dialect. I think it would be very interesting to test people who speak in a particular dialect, and then test another group of people who speak in a different dialect, and see whether you can find differences between Laurel and Yanny based on dialect. To my knowledge, that experiment hasn't been carried out yet.
Diana has conducted experiments using sonic illusions for years. Some even share similarities to this Laurel/Yanny debate.
Diana: Actually my phantom words illusion is very much related to the Laurel/Yanny illusion.
[SFX: phantom words]
Diana: It's very much related to the Laurel/Yanny thing, except that the Laurel/Yanny thing people are given what psychologists call forced choice. In my phantom words illusion they're not being asked whether or not they're hearing a particular word or a phrase. Instead, they're just told to say what it is that they hear, or to write down what it is that they hear, and that way you get many different answers, and that can be during listening to the same sequence of the identical words and phrases.
Ok, so let’s try this together… What do you hear? [SFX: phantom words]. I hear no way, no way, no way, no way, and I hear it in an American accent… but others on the team say they hear no where in a British accent [SFX: phantom words]. Here’s another example. [SFX: phantom words] I hear countdown, countdown, countdown, countdown… but others here on the team reported hearing the words Hilda, Hilda, Hilda, or Gilda, Gilda, Gilda, or Wando, Wando, Wando, or yoga, yoga, yoga, and even thank you, thank you, thank you. [SFX: continue example] What you hear seems to entirely depend on your language, your background, and your accent.
[SFX: continue example]
Diana: When we listen to speech, we construct for ourselves the words and the phrases. We don't really hear the actual sounds that are being spoken. We use our knowledge and our experience of sounds that are rather like different speech sounds to construct for ourselves the speech that is really being uttered. It's not surprising, therefore, that you can create illusions of speech deliberately that way. Because this process of construction goes on all the time when we're having conversations in everyday life.
We’ll hear even more sonic trickery… after the break.
Sonic illusions show us just how subjective sound can be. The Laurel/Yanny debate is just one example. Here’s Dana again deconstructing another viral sonic illusion.
Dana: Actually in the middle of a lab meeting, someone saw on Facebook or something this video of a little plastic toy that says either "Brainstorm," or "Green needle."
[SFX: Brainstorm/Green needle illusion]
Dana: We watched it several times, and what was interesting about that one is that it's a lot more what we call "cognitively penetrable." So you can sort of control what you're gonna hear by thinking about it a certain way, and you can flip back and forth depending on which thing you're thinking of. It's ambiguous and so there are lots of different ways that you could interpret it. And there's different groups of speech sounds that are similar in different ways, so you can definitely hear different combinations. Maybe hear green storm, or brain needle, or any other sorts of combinations of sounds. But the internet has a way of choosing two interpretations and pitting them against each other. Because I guess it just makes for more fun Twitter debates.
[SFX: Brainstorm/Green needle illusion]
Dana: I would say that a lot of illusions are more psychology, in that it's not the biology of the brain or the structure of the brain that is making these illusions possible. A lot of it is more in the higher level interpretation of the sensory information, or maybe your life experience or its certain assumptions that your brain is making.
Diana’s research validates that point. One of her earliest and best known audio illusions is called the Tritone Paradox. It consists of two computer-produced tones connected by a half-octave, called a tritone.
[SFX: Piano Tritone]
Diana: I can pretty much guarantee that listeners will disagree among themselves as to which tone pairs they hear as ascending, and which is descending.
So, I’m going to tell you what I hear. Here’s the first one...
[SFX: Tritone Paradox]
[Tritone 1] [Dallas mimics a low tone followed by higher tone]
[Tritone 2] [Dallas mimics a high tone followed by a lower tone]
[Tritone 3] [Dallas mimics a low tone followed by a higher tone]
[Tritone 4] [Dallas mimics a high tone followed by a lower tone]
Diana: At the time, I was teaching a group of students who were all Californians. They spoke Californian English, and their parents spoke Californian English. I’m from the South of England, I’m from London. I was very surprised to discover that I was hearing the opposite of what most of my students were hearing, and so it seemed to me that maybe language or dialect was an issue here. Then I went on a speaking tour to different places in Europe, and I found that different audiences had different flavors of what they were likely to hear.
When you think about it, it makes sense that where you grew up shapes how you interpret the sounds around you.
Dana: Your brain has to rely on its prior knowledge about the world and different assumptions that it makes about how the world works in order to figure out how to interpret this information that's coming in from the senses. And there's only, I would say, a smaller number of illusions that are actually due to the biology of how the cells in our sensory systems are structured.
But when biology does influence the illusion, it can be confusing. When Diana discovered her very first illusion, the octave illusion, she was baffled by her own experience.
Diana: I was experimenting with this software that would enable me to play two sequences of tones at the same time, one to my right ear, and the other to my left one.
[SFX: octave illusion]
Diana: It just became increasingly clear that something rather strange was happening. The pattern I devised consisted of two tones that were spaced an octave apart, and alternated repeatedly. The right ear received the sequence high tone, low tone, high tone, low tone over, [SFX: play sequence]. And at the same time, the left ear received low tone, high tone, low tone, high tone [SFX: play sequence] over, and over again. When I put on my earphones I was astonished. A single tone appeared to be switching back and forth, from ear to ear and at the same time, its pitch appeared to be switching back and forth from high to low.
So, if you have headphones on, here’s what I want you to do. Right now, reverse your headphones. So, put the left headphone on your right ear and the right headphone on your left ear. I’ll give you a few seconds to do this [SFX: elevator music for 5-10 seconds] Ok, here’s the illusion. While it’s playing I want you to determine which ear you’re hearing a high consistent tone. It’s this tone [SFX: Whistles]. There’s also a low consistent tone [SFX: Whistle]. One of the tones will be on one side, so your left or your right, and the other tone should be on the opposite side. Here we go. [SFX: octave illusion] Ok, remember which ear you heard the tones from. Now put your headphones back on normally. I’ll whistle the tone while we wait [SFX: whistling]. Ok ready, here it is again [SFX: octave illusion]. Did you hear the high and low tones in the same ears? Weird right?
[SFX: octave illusion]
No matter how the headphones are placed, Diana’s research eventually revealed that most right-handers hear the high tone on the right side. Left-handed and ambidextrous people are more varied in terms of where the high and low tones appear to be coming from.
[SFX: octave illusion]
A quick warning for those driving right now. The next minute has highway sounds like honks, sirens, and crashing.
[SFX: busy city street sounds etc.]
Dana: Our senses are our brain's only way of gathering information about the world out there and then using that information to take appropriate actions. But your brain doesn't actually just receive this information in a passive way from your eyes and your ears and other sensory organs. It gets raw information and then it has to interpret this information and actually create your perception of the world. So your brain has to decide what information is important, and what information can be discounted as just noise that's distorting the signal. And then it has to take this incomplete information and fill in the gaps as best it can, and make an inference or it's best guess about what's actually out there in the world. [SFX: Screeching tires and car crash sound]. [SFX: Siren] And this means that sometimes, your brain gets this wrong and illusions are a good example of when your brain gets this wrong, or it creates an image or a sound that actually isn't even there in the first place.
Dana uses FMRIs which allow her to look at the brain. Specifically the auditory cortex.
Dana: So the part of the brain that processes sound, to try to learn more about how it's organized. We know that the first place that sound goes in the cortex, which is the final processing stage after it comes up from our ears and through our brainstem. We know that the first place it goes in the brain is organized by frequency, almost like a piano from high to low, and then actually back to high. And that is mostly inherited from the way our cochlea is laid out, which is also by frequency.
Your cochlea is the spiral cavity of the inner ear. It kind of looks like a snail shell. It produces nerve impulses in response to sound vibrations.
Dana: But once the frequency content of the sound we're listening to has been figured out, there's a lot of other processing stages that happen in the auditory cortex. And we don't know much about what those processes are, whether they are different regions that are responsible for different kinds of sounds. We're pretty clear that there's a part of the brain that processes speech and just speech, but some people in my lab have found that there also seems to be a neural population part of the brain that cares a lot about music.
Dana: I just think it's exciting to be able to, look at the brain and try to actually understand it. It's still pretty cool when I'm scanning a subject, the image of their brain comes up on the computer screen and it's still cool to me that, that's them. That's their brain, that's how they're able to perceive and understand the world, and we're able to use math and physics to understand it.
If sonic illusions teach us anything, its that our hearing is a personal experience. Our lives shape the way we hear and react to sound. These illusions can affect us on an emotional level and help us understand that we live in a highly subjective reality. As we move through the world, we experience it – and process it -- in our own unique way.
Twenty Thousand Hertz is produced out of the studios of Defacto Sound, a sound design team that makes commercials, documentaries, and trailers sound incredible. To hear some of this sonic goodness, visit DefactoSound.com.
This episode was written and produced by Carolyn McCulley. And me, Dallas Taylor. With help from Sam Schneble. It was edited and sound design by Soren Begin. It was mixed by Jai Berger.
Thanks to our guests, Professor of Psychology at University of California San Diego, Dr. Diana Deutsch. And Dana Boebinger, a Ph.D. student at Harvard and MIT studying auditory cognitive neuroscience. The music is this episode is from our friends at MusicBed. Go listen at MusicBed.com.
You can check out our beautiful show art, and find full transcripts on our website - 20k dot org. And you can chat with me and the rest of the 20k team on Facebook, Twitter, or by writing hi at 20k dot org.
Thanks for listening!