6
$\begingroup$

Consider a source of sound such as a person speaking or a party of people which makes a continual drone sound of the the same frequency. If a human shakes their head side-to-side with sufficient angular speed, they are in effect obtaining different frequencies of the same sound source and should be able to apply the Doppler effect to approximately localize (from prior experience) the sound source.

Do humans use the Doppler effect to localize sources of sound and have there been any studies proving this?

Edit: A link to the Weber-Fechner law and a link to the wiki article discussing the just-noticable-difference (JND) for music applications were added to the OP for reference, based on the accepted answer.

$\endgroup$
4
  • 1
    $\begingroup$ Note: we will localize it quite effectively based on frequency responses of the ear, which differ from direction to direction. $\endgroup$ – Cort Ammon Mar 15 at 4:26
  • 2
    $\begingroup$ Expounding on @CortAmmon 's comment, the unique shape and ridges in our ears help to shape how our ears receive different frequencies. If you were to use clay to mold a smooth shape to your outer ears, you would lose the ability to localise sound. As demonstrated in a youtube video experiment performed on his own son by SmarterEveryDay. $\endgroup$ – Glen Yates Mar 15 at 14:16
  • 2
    $\begingroup$ @GlenYates More precise to say that the ability to localize sound is compromised by such a modification - but it is not fully lost. Modifying the shape of the ear only attacks one of our many mechanisms used to localize sound. It's an important one, but it is certainly not the only one. $\endgroup$ – J... Mar 17 at 13:32
  • 2
    $\begingroup$ @J... True, I forgot about localising left vs right by the time delta that sound is received among other things. We apparently use frequency analysis that is helped shaped by the shape of our ears to localise sounds that would otherwise be identical in time between left and right ears - such as whether a sound came from in front of you or behind you. It was this ability that was lost with somewhat comical effect in the mentioned video. $\endgroup$ – Glen Yates Mar 17 at 14:23
16
$\begingroup$

A person would not be able to localize a sound using the Doppler effect created by shaking their head.

Say a person shakes their head at 20 cm/s. The speed of sound is about 330 m/s. This gives a frequency change of 0.06%.

The "just noticeable difference" to discern two frequencies played in succession is about 0.6% (source), so about an order of magnitude too coarse.

$\endgroup$
12
  • 2
    $\begingroup$ @NiharKarve: Thanks. I've added a link, where 0.6% is the first sentence of the second paragraph under Music Production Applications. $\endgroup$ – tom10 Mar 15 at 3:53
  • 5
    $\begingroup$ ...but coherent doppler sensors usually measure a phase shift, not a frequency shift. (So it's the change in position, not the speed, that creates the signal). $\endgroup$ – JEB Mar 15 at 5:50
  • 1
    $\begingroup$ Running the numbers in my head, that suggests the just-noticeable difference is achieved by someone walking past (3.6km/h walking pace = 1 m/s = 0.3% of 330m/s, double it for approaching/receding). That feels right to me: I used to work with a part-time firefighter. When his emergency phone went off he sprinted out of work with it still sounding, and the shift was very clear (the ringtone was a two-tone siren - people got out of his way) $\endgroup$ – Chris H Mar 15 at 10:42
  • 1
    $\begingroup$ Sounds are affected by head angle anyway, so an erroneous signal could be perceived with shaking, but fast vs. slow shaking or indeed holding the extreme positions of the shake doesn't affect the pitch. As the ears are so near the axis of rotation, the linear motion is rather small. To get 20cm/s would seem to need very fast shaking indeed $\endgroup$ – Chris H Mar 15 at 10:45
  • 2
    $\begingroup$ We don't use the doppler effect, but the brain does cue off of the time delay between a sound reaching either ear. A sound coming from your left will hit the left eardrum almost a millisecond before the right. The brain can sense this difference and "moves" the sound to the left of your virtual soundstage in response. This can even be simulated with time delays in headphones where both sounds are otherwise identical (ie: no filtering, same volume, etc) - to distinguish from other cues like your head blocking high frequencies, etc. $\endgroup$ – J... Mar 15 at 13:26
1
$\begingroup$

Humans DO use dopler effect to estimate a sound source position, they just dont use it exactly the way you imagine.

The simplest example is the distance to a passing by object (a car, an airplane, a mosqito or even a talking human). A near flyby makes a rapidly lowering tone. An object passing away from you will change its tone slower.

Moving your head left and right gets you a direction by the phase difference between ears. Well, you can get the direction withour moving your head, but it will be exact up to the symmetry of your head.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.