[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: HC selectivity ... was Re: Physiological models of cochlea activity - alternatives to the travelling wave

On Mon, 1 Oct 2007, Martin Braun wrote:

> I understand the quoted part from the Wikipedia article "Scientific
> Modeling" without any mathematical formulae, even though I have not written
> it and even though I do not even know who has written it.

Of course, we are talking here about understanding natural phenomena
(dynamics of a complex system, to cite the original sentence), and not
texts. What do you mean by the understanding of a system? For many of us
this means being able to crate formulae that describe its behaviour
(they fit the data and are able to predict further data).

> You need meta-knowledge to be able to asses if a model is correct and
> useful. In science this meta-knowledge is: All Relevant Data.

We agree on that.

> I reject all models of cochlear mechanics that state that the BM triggers
> the OHCs at ALL sound levels. However beautiful these models are!

OK, I understand now. You are not willing to accept models that obviously
give only poor approximations. I think that sometimes they can still be
good -- if one always keeps in mind that they are just bad approximations
that need continuous refinement. But they indeed hold the risk the people
are inclined to forget this fact, and the model becomes "the master" and
not "the slave". In fact, I think that something like that is going on
in my field (automatic speech recognition), so I definitely understand
your point. (I don't know much about cochlear modelling, so I don't want
to comment on the current state of models in that area. I wanted to
reflect only on a more general level...)

               Laszlo Toth
        Hungarian Academy of Sciences         *
  Research Group on Artificial Intelligence   *   "Failure only begins
     e-mail: tothl@xxxxxxxxxxxxxxx            *    when you stop trying"
     http://www.inf.u-szeged.hu/~tothl        *