Hello Dick,
According to Chapter 10 of "Signals, Sound, and Sensation" by W. M. Hartmann (Springer, 1998), the widths of "Cambridge critical bands" are equal to "equivalent rectangular bandwidths", and these are (mainly?) measured by notched-noise experiments. I believe that the widths presented, e.g., in Hartmann's Fig. 10.9 are valid for soft tones. During a soft sine-tone of given frequency (above ~1 kHz), there is a high "active peak" in the curve of basilar-membrane oscillation velocity amplitude versus distance from stapes. See, e.g., my recent paper "Old and New Cochlear Maps", Canadian Acoustics Vol. 37, No. 3 (2009) 174-175. If one considers a fixed place on the basal half of the basilar membrane (i.e., a given distance from stapes), then there is a corresponding active peak in the curve of basilar-membrane oscillation velocity amplitude versus sine-tone frequency. The critical bands coincide, on the frequency scale, with the widths of those active peaks. So, the just described critical bands appear to be defined (mainly?) by basilar-membrane oscillation patterns.
With best wishes,
Reinhart.
----Ursprüngliche Nachricht-------------------------------
Von: DickLyon@xxxxxxx
Datum: 15.06.2010 03:45
An: <AUDITORY@xxxxxxxxxxxxxxx>
Betreff: Re: Rationale for Critical Bands
Dan, the situation is indeed complicated, but I don't think there's
much actual disagreement. [...]
------------------------------------------------
Reinhart Frosch,
Dr. phil. nat.,
r. PSI and ETH Zurich,
Sommerhaldenstr. 5B,
CH-5200 Brugg.
Phone: 0041 56 441 77 72.
Mobile: 0041 79 754 30 32.
E-mail: reinifrosch@xxxxxxxxxx .