Subject: Re: Is there considerable phase locking up to 6 kHz? From: "Richard F. Lyon" <DickLyon(at)ACM.ORG> Date: Wed, 17 Mar 2004 18:40:04 -0800At 10:44 PM +0100 03/17/2004, Martin Braun wrote: >We see two main pathways on which period information can reach the period >detectors in the midbrain: > >a) the proper pitch pathway, which delivers period coding, by phase locking, >of resolved harmonics, and > >b) a secondary pathway, which delivers temporal delay signals that were >originally encoded for the purpose of sound localization. Martin, thanks for your comments. I will be more than happy to adopt your suggestion and give up the old idea of "place pitch" as unsupportable by any evidence. However, I'm not sure I like your division of timing pathways. The first one (a), with "phase locking of resolved harmonics" seems to be rather narrowly construed. Sound does not in general consistent of discrete sinusoids, since it is not in general periodic, so the mechanisms must be much more general. Even for periodic sounds, partials are in general only partly "resolved". The phase locking in the auditory nerve will follow the time structure of arbitrary waveforms at the different points along the cochlea. Conceptualizing in terms of sinusoids does more harm then good, in my opinion. The second pathway (b) seems funny, too. Why would information be encoded "for the purpose of sound localization" and then used for a different purpose? Anyway, I agree with you that the temporal structure is available on the auditory nerve, and is used by binaural and pitch processing. Dick