Question about sound direction detection with human hearing.

[MrWizzard]

If you close your eyes and someone makes a sound, you are able to tell where it is. But how are you able to tell weather the sound is in front of you or behind you if the sound is the exact same volume and centered in front of your head or behind.

I can still tell if it's behind me or in front of me, anyone know how this works? Shouldn?t you be able to trick the brain if you do it just right?

 

[Mark R]

The outer ear (pinna) filters the sound differently depending on whether the sound comes from the front or behind (essentially like different settings on a graphic equaliser).

So, if your brain is expecting a certain type of sound, it is possible for it to work out whether it came from in front or behind depending on how well it matches.

With signal processing techniques, it is possible to filter sounds so that they sound as if they originated behind or in front of you - although to reproduce these really needs headphones.

 

[oynaz]

That is the reason why your is the shape it is, btw.

 

[dead1ne]

While it's been awhile since I've looked at this specifically I'm pretty sure that you can't really tell where the sound is coming from(infront or behind) to compensate for this you subconsciously turn your head a very small amount which allows you to pinpoint where the sound is coming from.

 

[Gibsons]

hrm... so if we put someone in an anechoic chamber, bolt their head down so they can't move it at all and produce a pure tone (say, 1khz) directly in front of or behind them... could they tell which direction it's coming from?

 

[d3n]

I think it also might have to do with a person?s ability to process that type of information in the brain. With a very exact setup hardly anyone would be able to pinpoint it without turning an ear, but some with very fast neural responses would have more of a chance to isolate the doppler effect of any sounds. A person turns their head to increase the effect and that is when they are best able to determine direction. A pure tone would make it much more difficult I would imagine.

 

[Mark R]

Originally posted by: Gibsons
hrm... so if we put someone in an anechoic chamber, bolt their head down so they can't move it at all and produce a pure tone (say, 1khz) directly in front of or behind them... could they tell which direction it's coming from?

Probably not. The only mechanism available for detecing front/back and up/down sound direction is the spectral pattern - a pure tone doesn't help.

Pure tones are very difficult to localise anyway: the only useful filtering effect is from the head shadow, and the between-ear phase shift isn't particularly useful because of range ambiguity (this is worse as frequencies get higher). Once you add echos and multi-path then interference fringing becomes a problem.

Broadband sounds are much easier to localise due to less ambiguity from the between-ear correlation, and a spectrum that potentially provides a lot of data about filtration. This important observation is used in modern sirens on emergency vehicles - the relatively narrow-band wailing sound is difficult to localise, so modern sirens intersperse it with bursts of white noise.

 

[Jeff7]

I seem to recall reading that human ears are not arranged perfectly symmetrically on the head, that one is slightly farther back, or higher than the other one, so one ear receives the sound at a slightly different time than the other, which aids us in determining where the sound is. I think I read it in some of the technical writings that came with Aureal Vortex2 soundcards, which is how they figured out how to make good positional audio from just two audio sources.

 

[eelw]

Saw on the Discovery Channel, a Japanese group experimenting with sounds to remotely control a human. So the subject would be wearing a pair of headphones. As the subject walked forward, they would play a sound that would cause the person to turn either left or right. It worked because the sound caused the body to feel as if it was dizzy and the body instinctively turned the appropriate direction to maintain balance.

 

[videogames101]

Originally posted by: eelw
Saw on the Discovery Channel, a Japanese group experimenting with sounds to remotely control a human. So the subject would be wearing a pair of headphones. As the subject walked forward, they would play a sound that would cause the person to turn either left or right. It worked because the sound caused the body to feel as if it was dizzy and the body instinctively turned the appropriate direction to maintain balance.

Thats interesting, mind control FTW, I'd like to read up on that, do you have a link or something?

 

[Kyanzes]

Originally posted by: MrWizzard
If you close your eyes and someone makes a sound, you are able to tell where it is. But how are you able to tell weather the sound is in front of you or behind you if the sound is the exact same volume and centered in front of your head or behind.

I can still tell if it's behind me or in front of me, anyone know how this works? Shouldn?t you be able to trick the brain if you do it just right?

You probably couldn't if you had no outer ear.

 

[eelw]

Originally posted by: videogames101

Originally posted by: eelw
Saw on the Discovery Channel, a Japanese group experimenting with sounds to remotely control a human. So the subject would be wearing a pair of headphones. As the subject walked forward, they would play a sound that would cause the person to turn either left or right. It worked because the sound caused the body to feel as if it was dizzy and the body instinctively turned the appropriate direction to maintain balance.

Thats interesting, mind control FTW, I'd like to read up on that, do you have a link or something?

I'll attempt to find it. It was on a segment I saw on the Discovery Channel.