Subject: Re: Autocorrelation From: Alexandra Hettergott <a.hettergott(at)WANADOO.FR> Date: Sun, 16 Jul 2000 11:03:23 +0200
Peter Cariani <peter(at)epl.meei.harvard.edu> wrote (in part) : >The alternative general hypothesis is that frequency representation is >based not on spatial profiles of filter activation, but by an analysis >of the temporal patterns that are produced by the filters. The different >tunings are then not the vehicle for fine frequency discrimination, but >the means by which multiple frequencies/periodicities/auditory objects >can be simultaneously represented [...] Humble me (and re music) again, just to briefly state that a) I do consider your above hypothesis plausible also with regard to timbral distinctability (by way of envelope-related temporal patterns), as e.g., in different simultaneously playing musical instruments ; b) -- since you were asking about "our" overall views concerning pitches of resolved and unresolved harmonics : (modestly) I would not imagine two qualitatively different kinds of neural mechanisms but a single one rather, showing yet a change in resolvability becoming "disrupted" in both directions of very high/low periodicities, respectively (i.e., integrating/disintegrating) (in which respect, e.g., towards the "lower" end, as mentioned, I do see also some relation regarding the inversely related (ir)resolvability in low frequencies (becoming "pulse-like") vs. that of consecutive pulses (becoming "pitch-like", e.g., with about 20pps)) ; c) what you are describing as >the more intervals are produced by phase-locking to individual partials >(which depends on phase-locking), the better their representation in the >population-interval distribution and the greater their resolvability. does make sense to me in sort of different (periodic) "grids" to be applied. ..... Sincerely (sorry), Alexandra Hettergott.