Heart rate variability at limiting stationarity: evidence of neuro-cardiac control mechanisms operating at ultra-low frequencies.

This study considers the linkage of exogenously stimulated emotional stress with the neurogenic regulation of heart rate operating at very low frequencies. The objectives were three-fold: to consider the present evidence that such a linkage exists as a primary phenomenon; to compare the potential of a frequency-domain method and a time-domain method in revealing this phenomenon by characterizing heart rate variability (HRV) at frequencies of [0.0005...0.4] Hz and to design, implement and report a physiological experiment in which alternating periods of exposure to bland and high valence visual stimuli might reveal this phenomenon. A methodical challenge was to optimize the length of exposure to the stimulus such that subjects did not have time to habituate to stimuli, whilst acquiring sufficient data (heart beats) such that the ultra-low frequency (ULF) components of HRV could be described. With exposure times set to approximately 5 min, during which time the strength of the stimulus and the corresponding evoked response were considered stationary, the lowest HRV frequency component that could be characterized was 0.003 Hz. In trials with parametrically defined test data, the time-domain method based on the Ornstein­Uhlenbeck Gaussian process (OU-GP) was shown to be better than the frequency-domain method in describing the ULF components of the HRV. In an experimental cohort of 16 subjects, analysis using the OU-GP revealed evidence of cardiac regulatory mechanisms influenced by emotional valence operating in the bandwidth (ULF*) [0.002...0.01] Hz.