I've recently seen the following image:

enter image description here

Source: https://commons.wikimedia.org/wiki/File:Hoerflaeche.svg

Translation (from German):

  • The upper boundary (Schmerzgrenze) is where it starts to hurt
  • "Musikwahrnehmbarkeit" is where you can hear music
  • "Sprachwahrnehmbarkeit" is where you can recognize speech
  • "Hörschwelle" is absolute threshold of hearing (ATH)

While the ATH seems to be relatively smooth, there is a noticeable "gap" between 2kHz and 5kHz. Does anybody know why this is the case?

Some possible explanations I've heard are:

  • Evolution: Hearing a baby cry
  • Evolution: Hearing leafs rustling
  • Coincidence: There is no good reason for that.

The image is from wikipedia commons and has no references. It might be wrong. Do you know similar graphs? Do you eventually have this kind of information for other animals?

  • $\begingroup$ The main reason is this is the area where harmonics in human speech are most prominent. $\endgroup$ – stix Jan 8 at 17:28

Acoustically, we hear better in the 2-5 kHz range because our auditory canal resonates in that range (specifically, around 3 kHz). Our ear looks something like this (U Miami):


The auditory canal is part of the outer ear, which amplifies sound so that it can be converted into electrochemical signals by the middle and inner ear. We hear better in the 2-5 kHz range because the resonant frequency of the auditory canal is in that range.

The mechanical question of "how does this work" is usually easier to answer than the "why did this evolve?" question. Since reproductive fitness is the driving force of evolution, hearing babies crying (between 1-5 kHz) is the best guess I can come up with.

  • $\begingroup$ Do you think the resonant frequency could be different without severely influencing other properties how we hear / what we hear? Is this different for other animals with similar ears? $\endgroup$ – Martin Thoma Dec 21 '14 at 21:26
  • 1
    $\begingroup$ Changing the resonant frequency wouldn't change what we hear but would change the threshold at which we hear different things. I don't actually know about other animals, perhaps ask another question if you want more information? $\endgroup$ – Luigi Dec 21 '14 at 21:29
  • $\begingroup$ This doesn't really answer OP's question, it only gives the mechanics behind it, and almost puts effect before cause. The reason the auditory canal is tuned to that frequency range is because our voices' harmonics are most prominent in that range. $\endgroup$ – stix Jan 10 at 4:38

You may never see this, but I also had this question - I think this may be a fun take.

The 2-5kHz range is associated with the third and fourth formants of the human voice, and these formants are used primarily in singing.1 After reading some discussion about possible evolutionary reasons that humans sing - It seems that sexual selection could have played a role2,driving improvements in perception of that frequency range.

Also, as mentioned earlier in this thread - the ear's resonant frequency is most sensitive around 3khz, which correlates nicely to the f3, f4 frequency range.

"... our auditory canal resonates in that range (specifically, around 3 kHz)." - Luigi

Finally, I'll link a neat video 3 in which a primatologist explains a phenomena in which female gorillas sing to emanate comfort/safety. Could possibly have a related function in humans.

p.s. please excuse the 3rd source :)

1^ Frequency range of human singing formants f3,f4

2^ A thread discussing the evolution of singing

3^ Gorillas singing analysis

  • $\begingroup$ This is exactly why, except that it's not just for singing. Human speech itself has a lot of energy in these bands. We're most sensitive in this region because it helps us to hear each other speak better. A lot of voice CODECS specifically oversample this region as a result. $\endgroup$ – stix Jan 8 at 17:30
  • $\begingroup$ @stix on the other hand, why didn’t the human speech/voice cord evolve to be in a different frequency band? $\endgroup$ – Kal Jan 9 at 2:44
  • 1
    $\begingroup$ @kal The average human trachea is around 11cm. The wavelength of 3 kHz is around 11 cm. That band is the resonant frequency of the average human vocal tract. It's due to the size our bodies evolved to be. $\endgroup$ – stix Jan 10 at 4:34
  • $\begingroup$ @stix thanks for verifying/clarifying :) $\endgroup$ – William Arnold Jan 13 at 22:03

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.