It has been hypothesized that obesity is an abnormally high setting of a regulatory system controlling the level of fat storage in adipose tissue (1). The regulation is achieved by the operation of many elements and is determined by genetic variation and environmental factors. The autonomic nervous system (ANS), 1 with both sympathetic and parasympathetic controls (SC and PC, respectively), is believed to play a particularly important role. SC has been intensively studied in obesity; in fact one theory of obesity causation is the Mona Lisa Hypothesis, an acronym for “most obesities known are low in sympathetic activity”(2). A widely used technique for evaluating SC in humans and in experimental animals is dependent on the turnover rate of norepinephrine as measured by the decline in plasma norepinephrine specific radioactivity after injection of tritiated norepinephrine (3). In experimental animals, direct measure of turnover in heart muscle has been a valuable tool. The most direct method for measuring SC is microneurography, using an electrode in the peroneal nerve (4). PC is also likely to be a player in fat storage, as suggested by the profound effect of vagotomy in ameliorating the effects of ventromedial hypothalamic lesioning. Some vagal effects may be related to insulin release, but there are likely other more direct relations of PC to fat storage, such as the marked bradycardia—probably not insulin related—occurring with weight reduction. Variations in heart rate, in particular, the waxing and waning of heart rate with respiration, or so-called sinus arrhythmia, is an effect of PC and thus has been used to estimate the degree of PC. In 1981, a research group at the Harvard–Massachusetts Institute of Technology Division of Health Sciences and Technology and the Massachusetts General Hospital published findings on formal spectral analyses of heart rate fluctuations to evaluate PC and SC (5). Modern electronic and computer technology enable rapid analysis of the “power” of various fluctuating or periodic inputs to heart rate at frequencies from 0 to 0.5 Hz (0 to 30 cycles/min). High-frequency peaks at 0.4 Hz are the result of sinus arrhythmia. Because sinus arrhythmia can be completely alleviated by atropine, a muscarinic parasympathetic blocker, its measure is believed to be a “pure” measure of the parasympathetic input to heart rate variability at the frequency of respiration. Lower frequency peaks were found to be a mixture of SC and PC, and another input from the renin–angiotensin system was also identified. The equipment to perform such analyses is now readily available and has often been used to evaluate SC and PC in humans, both obese and non-obese, under a variety of conditions. In this issue of Obesity Research, Nagai et al.(6) make use of this method to evaluate the relationship of ANS activity to the development of obesity in Japanese school children. We have studied spectral analyses of heart-rate variability in human subjects (7), hoping that this relatively simple and noninvasive method would give reliable measures of SC and PC in human subjects as levels of fat storage are modified by feeding low or high calorie diets. For the reasons given below, we have found that the method has several shortcomings, and a more complex approach using pharmacological blockade becomes necessary for accurate assessment of ANS (8).

The beat-to-beat interval (RR) in milliseconds is shown in the Nagai et al. study (6) of autonomic activity in obese and non-obese Japanese children. They illustrate the variation of heart rate in the “time domain,” and the same figure shows how these data can be reanalyzed in the “frequency domain …