Principles and applications of fluctuation analysis: a nonmathematical introduction.

The mechanisms underlying many of the processes studied by membrane biophysicists are inherently probabilistic, and therefore exhibit random fluctuations around the mean of behavior. These fluctuations reflect the underlying probabilistic mechanism and therefore can sometimes provide information, not otherwise available, about these mechanisms. Fluctuations may be characterized by their spectra which are obtained from a Fourier analysis of the experimental records. When a theory for membrane processes is available, it makes predictions about fluctuation spectra and therefore may be tested by examining these spectra. Theories about gating behavior at the frog neuromuscular junction have been tested in this way, and it has been possible, in addition, to estimate the conductance of one open channel, a quantity not susceptible to direct measurements. Various physical pictures are capable of yielding the same macroscopic behavior for axon membranes, that is, the Hodgkin-Huxley equations, but these various mechanisms predict that the current fluctuations around their mean values should have different characteristics. Fluctuation analysis may, then be of value in elucidating the physical basis for axon conductance changes.