Permeability is related to the ease with which an ion goes through
the membrane, either absolutely, or in the case of the GHK equation, ease
relative to other ions. So, yes, its ability to compete for channel
binding sites with other ions. Permeability can be measured by
radioactive flux experiments, or relative permeability by reversal
potential measurements. This term applies to single ions, and hence
permeability is not concentration-dependent, except in unusual
circumstances (e.g., ion concentration affects number of open channels).
Conductance measures how many ions per second cross the membrane, and
hence is measured as a current at a certain voltage. Conductance is
concentration-dependent, hence the use of high concentration of permeant
ions to measure single channels in patch clamp experiments.
If a channel selects by size, such as the Na channel, the
conductance and permeability of various ions through it fall in the same
order: a small, selected ion moves fast throught the channel. For
channels, such as the L-type Ca channel, which select by binding, the two
are inversely related. An ion with high relative permeability, such as
Ca over Na, moves slowly through the channel because it is bound by the
selectivity site. This is most easily seen with Na moving through the Ca
channel: in mixtures of Ca and Na, the channel's permeability to Ca over
Na is very high, and almost no Na moves through. If you remove all Ca
(and Mg), the Ca channel passes Na very rapidly, and Na currents through
Ca channels are much larger than Ca currents.
About 25 years ago, Alex Mauro wrote a beautiful article defining the
relationship between circuit and diffusional models of the membrane, and
if you are serioud about this, it is well worth wading through. e-mail
me if you want the citation, as I can't find it right now.