Estimates for ELF effects: Noise-based thresholds and the number of experimental conditions required for empirical searches

Abstract
Interactions between physical fields and biological systems present difficult conceptual problems. Complete biological systems, even isolated cells, are exceedingly complex. This argues against the pursuit of theoretical models, with the possible consequence that only experimental studies should be considered. In contrast, electromagnetic fields are well understood. Further, some subsystems of cells (viz. cell membranes) can be reasonably represented by physical models. This argues for the pursuit of theoretical models which quantitatively describe interactions of electromagnetic fields with that subsystem. Here we consider the hypothesis that electric fields, not magnetic fields, are the source of interactions, From this it follows that the cell membrane is a relevant subsystem, as the membrane is much more resistive than the intra‐ or extracellular regions. A general class of interactions is considered: electroconformational changes associated with the membrane. Expected results of such as approach include the dependence of the interaction on key parameters (e.g., cell size, field magnitude, frequency, and exposure time), constraints on threshold exposure conditions, and insight into how experiments might be designed. Further, because it is well established that strong and moderate electric fields interact significantly with cells, estimates of the extrapolated interaction for weaker fields can be sought. By employing signal‐to‐noise (S/N) ratio criteria, theoretical models can also be used to estimate threshold magnitudes. These estimates are particularly relevant to in vitro conditions, for which most biologically generated background fields are absent. Finally, we argue that if theoretical model predictions are unavailable to guide the selection of experimental conditions, an overwhelmingly large number of different conditions will be needed to find, establish, and characterize bioelectromagnetic effects in an empirical search. This is contrasted with well‐established chemical dosimetry, which is much simpler. Because of the large number of possible electromagnetic field conditions, we also conclude that in vitro studies, rather than in vivo studies, should be emphasized in studies aimed at discovering and characterizing mechanisms for bioelectromagnetic effects. 1992 Wiley‐Liss, Inc.