Abstract
Analysis of current recordings from acetylcholine-activated channels has largely rested so far on the hypothesis of independence, which states that the opening of one channel does not influence that of its neighbors. We have submitted this assumption to several tests, using as experimental material single channel currents from rat myotubes. We found that, even though the distribution of multiple channel openings may be approximated by the Poisson law, openings are strongly coupled. This conclusion is derived from the analysis of two-time properties associated with patch-clamp data. We show how these properties, which contain more detailed information than the stationary probabilities, can be calculated in practice and why a Poisson analysis is misleading in the present case. The implications of our findings are finally discussed in terms of channel structure and function.