2 Jan
2013
2 Jan
'13
9:43 a.m.
On Mon, Dec 31, 2012 at 5:01 PM, Folderol <folderol(a)ukfsn.org> wrote:
On Mon, 31 Dec 2012 15:39:45 -0600
Charles Henry <czhenry(a)gmail.com> wrote:
Complicated. To the best of my recollection...
The hair cells are half-wave rectifiers and trigger on the positive
zero crossing. They can fire, at fastest, about 1000 times a second.
So, at lower frequencies, the hardware is in fact transmitting a
mostly-phase-locked response to the incoming signal. This is part of
how low-midrange frequency sounds are localized.
Above 1000Hz the firing pattern becomes more like a power integral and
phase information is ~discarded. At low frequencies, it appears that
the neural hardware only pays attention to phase in the sense of
comparing phase offset between the ears. The transition is gradual.
So the ears hear absolute phase < ~ 1000Hz but it's not clear absolute
phase is actually passed up the perception chain.
However, phase shifts aren't much of a
problem by themselves--they're all
over the place in any typical multi-driver system and placement in the room
matters just as much. The loudspeaker crossover introduces the same effect.
A slight digression...
For a long time I've wondered if the ear, being a non-linear 'device' can
actually detect absolute phase at low frequencies. i.e. if the compressions and
rarefactions were swapped, would it sound different? To test this with a mono source presumably all
you'd have to do would be to
have an asymmetric signal and swap the speaker leads, but how would you
objectively test the listener?
Simple t-test should do it :-)
Monty