| Abstract: | Animal studies of the effects of low-frequency electromagnetic fields (EMFs) on the immune system appear inconsistent and recent evidence indicates that inconspicuous experimental problems are not responsible. We hypothesized that the inconsistencies resulted from use of linear methods and models to study inherently nonlinear input-output relationships. Using a novel analytical method, we found that exposure of mice to 5 G, 60 Hz, for 1–105 days in 6 independent experiments consistently affected a broad panel of immune variables when and only when the reaction of the immune system was modeled to allow the possibility of nonlinearity in the relationship between the field and the immune variables. It was possible to mimic the pattern observed in the immune data by sampling from a known chaotic system, suggesting the possibility that the observed pattern was the result of intrinsic nonlinear regulatory mechanisms in the immune system. Overall, the results suggested that lymphoid sub-populations were vulnerable to the physiological consequences of EMF transduction, that it may never be possible to predict specific changes in particular immune-system variables, and that the underlying behavior of the immune system (that which occurs in the absence of specific inputs) may be governed by laws that manifest extreme sensitivity to prior states. |
