Sounds just fine
April 10, 2006 5:18 AM Subscribe
If sound were to travel closer to, say, the speed of light than the speed of sound, would we still be able to determine its horizontal direction?
My understanding is that we determine where sound is coming from along the x-axis based on the difference in time between the sound hitting our two ears. The space between our ears is enough for our brain to differentiate and make a determination about location.
If higher-pitched sounds were to travel fast enough, I imagine we'd lose that ability. (1)True? (2)Is it possible to augment the speed of sound (atmospheric density is at least one part of this, yes?) to the point where we'd lose this ability?
(3)Why do lower frequency sounds like those generated by subwoofers and low bass notes already evade our ability to determine their direction? (4)And is there a term for this sort of sound?
(5)Will losing hearing completely in one ear render, as I suspect, you entire ability to locate sound useless? And finally, (6)do our brains need to recalibrate as the distance shifts in our earliest ears? Or does our locational hearing just suck for awhile until we get to a certain head size?
Help me wrap my head around this!
My understanding is that we determine where sound is coming from along the x-axis based on the difference in time between the sound hitting our two ears. The space between our ears is enough for our brain to differentiate and make a determination about location.
If higher-pitched sounds were to travel fast enough, I imagine we'd lose that ability. (1)True? (2)Is it possible to augment the speed of sound (atmospheric density is at least one part of this, yes?) to the point where we'd lose this ability?
(3)Why do lower frequency sounds like those generated by subwoofers and low bass notes already evade our ability to determine their direction? (4)And is there a term for this sort of sound?
(5)Will losing hearing completely in one ear render, as I suspect, you entire ability to locate sound useless? And finally, (6)do our brains need to recalibrate as the distance shifts in our earliest ears? Or does our locational hearing just suck for awhile until we get to a certain head size?
Help me wrap my head around this!
Best answer: Sound that's below the range of human hearing is 'infrasonic'... above is 'ultrasonic'.
Directional hearing isn't just time, either. Here's an experiment you can try for yourself: fold the tops of your ears down so that they're more or less at a 90 degree eangle to their normal position. Dn't block the ear, just fold down the top flap 90 degrees. You will lose the ability to locate sounds in the vertical plane. I think you may lose it in the horizontal plane too, but I'm not sure and can't experiment right now. Try it, it's interesting.
Considering how badly a folded ear messes up sound location, drastic stuff like changing the speed of sound significantly will almost certainly break it, at least for awhile. The brain might be able to adapt.
posted by Malor at 5:50 AM on April 10, 2006
Directional hearing isn't just time, either. Here's an experiment you can try for yourself: fold the tops of your ears down so that they're more or less at a 90 degree eangle to their normal position. Dn't block the ear, just fold down the top flap 90 degrees. You will lose the ability to locate sounds in the vertical plane. I think you may lose it in the horizontal plane too, but I'm not sure and can't experiment right now. Try it, it's interesting.
Considering how badly a folded ear messes up sound location, drastic stuff like changing the speed of sound significantly will almost certainly break it, at least for awhile. The brain might be able to adapt.
posted by Malor at 5:50 AM on April 10, 2006
Regarding (5), I'm almost completely deaf in my right ear, and have a hell of a time knowing where sounds come from. It was hell in gradeschool sports.
posted by notsnot at 5:57 AM on April 10, 2006
posted by notsnot at 5:57 AM on April 10, 2006
Best answer: i think it's more complex than that - i suspect that the shape of the ears makes sounds from different directions sound slightly different too.
some initial numbers: sound travels at about 340 m/s. concert A is 440 Hz. so a single wavelength "lasts" for 1/440 s and has a length (how far the sound moves in that time) of 340 * 1/440. so you're talking about wavelengths of around 80cm. that's similar, but larger than the size of a head.
now you need to be clear about what "hitting our ears" means. that gives an image of sound like particles, hitting our ears at particular times. it's better to think of sound as continuous waves, and to compare the arrival of the "peak" of the wave (that may seem odd - a sound has to "start" - but it turns out to be somewhat equivalent - a starting sound includes high pitches with short wavelengths, so you get the extra accuracy you'd expect from listening to an abrupt "pulse" of sound using the same continuous wave model, because the higher frequencies have shorter wavelengths and so give you better resolution).
if the ears compare sound arriving at both sides of the head to determine the phase shift, you need to the distance between the ears to be a significant fraction of a wavelength (if it's much less than a wavelength your ears are hearing "unrelated" sounds; if it's a getting close to a wavelength, the difference is too small to hear, and larger than a wavelength you're again "unrelated").
i guess ears are about 20cm apart. and from the argument above, you'd expect this to work when the wavelength of sound is a bit bigger than that - so say for wavelengths from 40 to 80cm. if concert A is 80cm that's between that and an octave higher. that's pretty high, which bears out the idea that it won't work so well for a subwoofer.
in practice, again, i suspect a range for frequencies helps. you can imagine the brain doing some pretty complex processing, using different frequencies to remove ambiguities and extending the frequency range.
ok, and if the speed were different? then, assuming the frequency remained the same, the wavelength would get longer (if the speed gets higher - travels further in 1/440 s, for example). so if the speed doubled, you'd expect to have to double the frequency of the sounds that "worked". and if it went up by a factor of a million (which is what you're suggesting) then the frequency would need to go up that high too.
which sounds impossible. but if the physical world were such that the speed of sound were so high, then i suspect we would listen to frequencies that high too. my guess is that it's more the wavelength that's fixed (by the physical size of things). so my guess is that in an alternate universe with sound that travels at the speed of light this would still work, because we'd listen to correspondingly higher frequencies. but i'm not really sure it would be possible - everything would be very rigid/hard....
posted by andrew cooke at 6:02 AM on April 10, 2006
some initial numbers: sound travels at about 340 m/s. concert A is 440 Hz. so a single wavelength "lasts" for 1/440 s and has a length (how far the sound moves in that time) of 340 * 1/440. so you're talking about wavelengths of around 80cm. that's similar, but larger than the size of a head.
now you need to be clear about what "hitting our ears" means. that gives an image of sound like particles, hitting our ears at particular times. it's better to think of sound as continuous waves, and to compare the arrival of the "peak" of the wave (that may seem odd - a sound has to "start" - but it turns out to be somewhat equivalent - a starting sound includes high pitches with short wavelengths, so you get the extra accuracy you'd expect from listening to an abrupt "pulse" of sound using the same continuous wave model, because the higher frequencies have shorter wavelengths and so give you better resolution).
if the ears compare sound arriving at both sides of the head to determine the phase shift, you need to the distance between the ears to be a significant fraction of a wavelength (if it's much less than a wavelength your ears are hearing "unrelated" sounds; if it's a getting close to a wavelength, the difference is too small to hear, and larger than a wavelength you're again "unrelated").
i guess ears are about 20cm apart. and from the argument above, you'd expect this to work when the wavelength of sound is a bit bigger than that - so say for wavelengths from 40 to 80cm. if concert A is 80cm that's between that and an octave higher. that's pretty high, which bears out the idea that it won't work so well for a subwoofer.
in practice, again, i suspect a range for frequencies helps. you can imagine the brain doing some pretty complex processing, using different frequencies to remove ambiguities and extending the frequency range.
ok, and if the speed were different? then, assuming the frequency remained the same, the wavelength would get longer (if the speed gets higher - travels further in 1/440 s, for example). so if the speed doubled, you'd expect to have to double the frequency of the sounds that "worked". and if it went up by a factor of a million (which is what you're suggesting) then the frequency would need to go up that high too.
which sounds impossible. but if the physical world were such that the speed of sound were so high, then i suspect we would listen to frequencies that high too. my guess is that it's more the wavelength that's fixed (by the physical size of things). so my guess is that in an alternate universe with sound that travels at the speed of light this would still work, because we'd listen to correspondingly higher frequencies. but i'm not really sure it would be possible - everything would be very rigid/hard....
posted by andrew cooke at 6:02 AM on April 10, 2006
oh, and yes, i would guess that your head re-calibrates all the time. we're organic and adaptable - it's probably much "easier" to have something that is roughly stable and gets recalibrated all the time, than something that is completely fixed forever.
(ie there's probably a difference between doing this or not, and how accurately it's done. we probably learn to do it fairly early in development, or even have it "hard-wired", but the exact "calculation" is probably refined every day. but this is just a guess).
posted by andrew cooke at 6:05 AM on April 10, 2006
(ie there's probably a difference between doing this or not, and how accurately it's done. we probably learn to do it fairly early in development, or even have it "hard-wired", but the exact "calculation" is probably refined every day. but this is just a guess).
posted by andrew cooke at 6:05 AM on April 10, 2006
(6) Babies who are just a few months old will look towards something that's making a noise. If we don't have directional hearing at birth, we gain it very early. And yeah, that probably means we have to keep adjusting it as we grow.
(I can't find a source now, but I remember hearing about an experiment where the subjects wore goggles that inverted their field of vision. Within a day or two, their brains had compensated and they were seeing things right-side-up even though the goggles were still delivering upside-down images to their eyes. If this is true, then it suggests that our brains adjust very quickly when our sensory apparatus changes. So it's not unreasonable that a growing child could adjust to a bigger head and wider-spaced ears in a day or two.)
posted by nebulawindphone at 6:10 AM on April 10, 2006
(I can't find a source now, but I remember hearing about an experiment where the subjects wore goggles that inverted their field of vision. Within a day or two, their brains had compensated and they were seeing things right-side-up even though the goggles were still delivering upside-down images to their eyes. If this is true, then it suggests that our brains adjust very quickly when our sensory apparatus changes. So it's not unreasonable that a growing child could adjust to a bigger head and wider-spaced ears in a day or two.)
posted by nebulawindphone at 6:10 AM on April 10, 2006
My understanding is that we determine where sound is coming from along the x-axis based on the difference in time between the sound hitting our two ears.
That's part of directional hearing. But you seem to be overlooking the most obvious one: the difference in volume between the two ears. That's how stereo recording works; you want the sound to be picked up by both mics at the same time, but at different volumes.
Also, even if you can only hear in one ear, a sound coming from different directions will have different frequencies emphasized. For example, if you set up a directional mic in a room and have one person sit in front of it and one person sit behind it, you easily can tell which is which by listening. The voice from behind the microphone will bounce off the wall and then get picked up by the mic and will sound like it's been EQed.
posted by ludwig_van at 6:38 AM on April 10, 2006
That's part of directional hearing. But you seem to be overlooking the most obvious one: the difference in volume between the two ears. That's how stereo recording works; you want the sound to be picked up by both mics at the same time, but at different volumes.
Also, even if you can only hear in one ear, a sound coming from different directions will have different frequencies emphasized. For example, if you set up a directional mic in a room and have one person sit in front of it and one person sit behind it, you easily can tell which is which by listening. The voice from behind the microphone will bounce off the wall and then get picked up by the mic and will sound like it's been EQed.
posted by ludwig_van at 6:38 AM on April 10, 2006
Response by poster: Nebula: I've heard of the goggle's trick, too.
Malor: You nail it in line with HowStuffWork's explanation:
What about bass, though? It's not "infrasonic," it's just not directional. Why do lower frequencies cause us trouble?
posted by disillusioned at 6:42 AM on April 10, 2006
Malor: You nail it in line with HowStuffWork's explanation:
The pinna, the outer part of the ear, serves to "catch" the sound waves. Your outer ear is pointed forward and it has a number of curves. This structure helps you determine the direction of a sound. If a sound is coming from behind you or above you, it will bounce off the pinna in a different way than if it is coming from in front of you or below you. This sound reflection alters the pattern of the sound wave. Your brain recognizes distinctive patterns and determines whether the sound is in front of you, behind you, above you or below you.Andrew: Thanks for the hard numbers. That's what I was looking for.
What about bass, though? It's not "infrasonic," it's just not directional. Why do lower frequencies cause us trouble?
posted by disillusioned at 6:42 AM on April 10, 2006
What andrew cooke said (the first time.) The ability to locate sounds depends on the wavelength of the sound wave; and if the speed of sound were higher, the frequency needed to get that wavelength will also increase. The factor the frequency increases by should be (new speed of sound)/(old speed of sound), which would be something large like 104 if the new speed of sound were even 1% of the speed of light.
posted by Johnny Assay at 6:45 AM on April 10, 2006
posted by Johnny Assay at 6:45 AM on April 10, 2006
Best answer: re bass: because the wavelength is so long both ears are hearing almost the same thing.
posted by andrew cooke at 6:46 AM on April 10, 2006
posted by andrew cooke at 6:46 AM on April 10, 2006
(and the wavelength is so long that the shape of the ears can't affect it much either - it's a general rule that things interact with waves (in interesting/useful ways) only if they're about the same size: much smaller and they're ignored; much bigger and they're just a brick wall).
posted by andrew cooke at 6:51 AM on April 10, 2006
posted by andrew cooke at 6:51 AM on April 10, 2006
I wonder if bone conduction is also a factor. Bone conduction hearing is stronger for low frequencies, so you hear bass notes less "with your ears" and more "with your skull." Since less of the input is coming from your outer ears, that might make it harder to use the shape of your ears to tell where the sound's coming from, no?
posted by nebulawindphone at 6:55 AM on April 10, 2006
posted by nebulawindphone at 6:55 AM on April 10, 2006
Best answer: There are a few factors which allow you to locate sound in space, some already mentioned:
. intensity
. phase
. shape of ear
. absorption by the head
. movement of the head
. confirmation from other senses
In actuality the last two are the most important, although we're not always aware of using them. If you close your eyes, forget about the environment and locate sounds with your head fixed in one position, you won't be very good at it.
However, to support the answers above, the speed of sound (all other things being equal) has nothing to do with any of these. Phase is important, studies in sound and recreation of positional or binaural audio can recreate position by playing with the phase of a signal as it reaches each ear. The absorption characteristics of your head mean that there is a muffling as sound passes through it which is an important spatial clue.
Much more information here .
posted by grahamwell at 6:55 AM on April 10, 2006
. intensity
. phase
. shape of ear
. absorption by the head
. movement of the head
. confirmation from other senses
In actuality the last two are the most important, although we're not always aware of using them. If you close your eyes, forget about the environment and locate sounds with your head fixed in one position, you won't be very good at it.
However, to support the answers above, the speed of sound (all other things being equal) has nothing to do with any of these. Phase is important, studies in sound and recreation of positional or binaural audio can recreate position by playing with the phase of a signal as it reaches each ear. The absorption characteristics of your head mean that there is a muffling as sound passes through it which is an important spatial clue.
Much more information here .
posted by grahamwell at 6:55 AM on April 10, 2006
If sound were different, wouldn't we have evolved to hear it differently?
posted by Pollomacho at 8:11 AM on April 10, 2006
posted by Pollomacho at 8:11 AM on April 10, 2006
I think andrew cooke's estimate of the frequencies that we are most direction sensitive is a bit high, but otherwise he is right on the mark. I think the directional sense should work very well for spacing anywhere between 1/4 wavelength (or even a little less) upto a little more than 1 wavelength (possibly still effective at 3-5 wavelengths).
The maximum difference in pressure reading will occur at 1/2 wavelength (peak of wave to trough of wave), but I don't think you need to be at the maximum. The measurement might be reading the power level, rather than the absolute signal level (absolute value squared). If that is true then the maximum difference in the readings will happen at 1/4 wavelength (from a zero crossing to either peak). If you use the ability to determine a change in volume as a guide, the two readings only have to be about 1dB different, which is somewhat less than 1/8 wavelength. I admit, I am mixing analysis methods a little, and more rigor is called for..
Anyway, 20cm is 1/8th wavelength at about 200Hz, which is right in the meat of the highest amplitude output of a human voice..
posted by Chuckles at 10:49 AM on April 10, 2006
The maximum difference in pressure reading will occur at 1/2 wavelength (peak of wave to trough of wave), but I don't think you need to be at the maximum. The measurement might be reading the power level, rather than the absolute signal level (absolute value squared). If that is true then the maximum difference in the readings will happen at 1/4 wavelength (from a zero crossing to either peak). If you use the ability to determine a change in volume as a guide, the two readings only have to be about 1dB different, which is somewhat less than 1/8 wavelength. I admit, I am mixing analysis methods a little, and more rigor is called for..
Anyway, 20cm is 1/8th wavelength at about 200Hz, which is right in the meat of the highest amplitude output of a human voice..
posted by Chuckles at 10:49 AM on April 10, 2006
I think everyone has already pretty much answered the meat of this question, but I just want to point out that sound could never travel at the speed of light, because it is a physical (as opposed to electromagnetic) wave generated by pressure differentials.
posted by Rhomboid at 4:05 PM on April 10, 2006
posted by Rhomboid at 4:05 PM on April 10, 2006
rhomboid:
just to further digress, relativistic phonons in bose-einstein condensates. but yes - in normal matter, not possible. maybe in neutron stars, black holes or some other sufficiently-dense medium though? not an astronomer, don't know.
posted by sergeant sandwich at 3:49 PM on April 11, 2006
just to further digress, relativistic phonons in bose-einstein condensates. but yes - in normal matter, not possible. maybe in neutron stars, black holes or some other sufficiently-dense medium though? not an astronomer, don't know.
posted by sergeant sandwich at 3:49 PM on April 11, 2006
« Older How can I work out which parts of my yard are... | Homeschooling info and related background check... Newer »
This thread is closed to new comments.
posted by 517 at 5:43 AM on April 10, 2006