Subject: Re: HC selectivity ... was Re: Physiological models of cochlea activity - alternatives to the travelling wave From: "Richard F. Lyon" <DickLyon@xxxxxxxx> Date: Fri, 28 Sep 2007 08:12:43 -0700 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>At 4:16 PM +0200 9/28/07, Toth Laszlo wrote: >According to its scientific definition, a good model "must accurately >describe a large class of observations [..] and it must make definite >predictions about the results of future observations" (S. Hawking). >In that sense the alchemists' speculations could hardly be called models, >... Actually, even the Alchemists' models, and early chemical models that caused Priestley to refer to Oxygen as "dephlogisticated air", were useful attempts at organizing the knowledge of that time. They weren't "good" models by Hawking's definition, but they were the best they had come up with so far. Eventually they were replaced by better model's (like Lavoisier's) that did a better job of explaining the observations. One has to formulate the models in order to test the ideas, see if they work, form better ideas and better models, etc. I think the cochlear models that many people use are similarly not "good"; but better ones do exist, and we need to continue to test them, to see what they explain and predict, do see which ones are best. Martin has his own models, but he doesn't want to call them that. Dick