Application of the maximum-entropy method to data on short-term variability in inter-beat interval and an interpretation of the calculated powers

Abstract

Sixteen young males underwent resting and mental arithmetic and were injected with 100 μg phenylephrine during the interventions. Half of them further underwent trinitroglycerine injections. Power spectral density for inter-beat interval variability was calculated using the maximum-entropy method (MEM) with a very high lag, and its validity was checked by nonlinear least-squares method. It was found that the MEM could produce a very accurate power spectral density. Results on the low- and high-frequency (LF and HF) powers indicated that the phenylephrine injection increased both the LF and HF powers, that the trinitroglycerine injection increased the LF power to some degree, and that the mental arithmetic decreased both the LF and HF powers with both phenylephrine and trinitroglycerine. Therefore, the LF power could be interpreted as a marker of both the vagal and sympathetic activities, whereas the HF power seemed to be solely associated with vagal activity. The implications of the interpretation are discussed in terms of the controversy on the LF power (vagal plus sympathetic vs. sympathetic). © 1998 Japanese Psychological Association. Published by Blackwell Publishers Ltd.